Early warning scoring system and method for high-risk newborns
Patent Information
- Application Number
- CN202411241400.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-05
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-09-05
Smart Images

Figure CN118762837B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of medical early warning, and in particular to a high-risk newborn early warning scoring system and method. Background Art
[0002] In the field of medical health monitoring and early warning systems, especially in the field of neonatal intensive care, this field focuses on real-time monitoring and management of the vital signs and physiological status of newborns in the neonatal intensive care environment. Specifically, the monitoring and early warning of high-risk newborns involves high-precision continuous monitoring of the personality characteristics and physiological status of newborns to detect potential health threats in a timely manner. However, since the physiological status of newborns, especially premature infants, is very sensitive, abnormal changes in physiological parameters often occur during nursing interventions, which poses a challenge to clinical judgment. Nursing interventions include position changes, suctioning, and feeding.
[0003] At present, in the process of neonatal care, it is often difficult to accurately distinguish the fluctuations of physiological parameters caused by nursing interventions from actual health risks. The difficulty of this distinction lies mainly in the fact that nursing interventions can cause physiological changes similar to health problems, such as rapid changes in heart rate and respiratory rate, or fluctuations in blood oxygen saturation. Existing monitoring systems usually have difficulty in making accurate distinctions between these fluctuations, which may lead to false alarms or missed alarms, thus affecting the judgment of medical staff and even delaying timely intervention for newborns. Summary of the invention
[0004] In view of the deficiencies of the prior art, the present invention provides a high-risk neonatal early warning scoring system and method to solve the technical problem that the medical monitoring system in the above-mentioned background technology is prone to false alarms or missed alarms.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions:
[0006] In a first aspect, the present invention provides a high-risk neonatal early warning scoring system, comprising: a status monitoring module, a data preprocessing module, an early warning control module, a preliminary identification module, an abnormal analysis module and a scoring module;
[0007] The state monitoring module is used to monitor the newborns in the neonatal intensive care unit in real time to obtain relevant personality characteristic data information and relevant physiological state data information of the newborns;
[0008] The data preprocessing module is used to preprocess the relevant personality characteristic data information and the relevant physiological status data information, remove noise, smooth the image and perform geometric correction, generate a monitoring data set, and use dimensionless processing technology to standardize the relevant data information in the monitoring data set;
[0009] The warning control module is used to construct a warning degree factor Ycyz based on relevant personality characteristic data information, and set a first warning threshold according to the value of the warning degree factor Ycyz. ;
[0010] The preliminary identification module is used based on the relevant physiological status data information of the newborn and combined with the first warning threshold , identify abnormal physiological parameters and issue preliminary warning instructions;
[0011] The abnormal analysis module is used to perform statistics on abnormal physiological parameters after receiving the preliminary warning instruction to obtain an abnormal indicator set, and construct a warning score index Ypzs according to the abnormal indicator set;
[0012] The scoring module is used to pre-set an evaluation threshold, and compare and analyze the evaluation threshold with the early warning scoring index Ypzs to comprehensively determine the current dangerous state of the newborn and take corresponding intervention measures according to the dangerous state.
[0013] Preferably, the state monitoring module includes a first data acquisition unit and a second data acquisition unit;
[0014] The first data acquisition unit is used to monitor and record relevant personality characteristic data information of newborns in the neonatal intensive care unit in real time, wherein the relevant personality characteristic data information includes the birth weight Ctz and gestational age Tz of several newborns in the neonatal intensive care unit;
[0015] The No. 2 data acquisition unit is used to monitor and record relevant physiological status data information of the newborn in real time, wherein the relevant physiological status data information includes the heart rate parameters, blood pressure parameters, respiratory rate parameters, blood oxygen saturation parameters, body temperature parameters, blood sugar level and cardiac output Xsc of the newborn in each monitoring period under nursing intervention.
[0016] Preferably, the data preprocessing module includes a processing unit and a unifying unit;
[0017] The processing unit is used to identify and filter out noise in the relevant personality characteristic data information and the relevant physiological status data information by using a filter, and smooth them by using a bilateral filtering method, and fill in missing values, wherein the missing value filling methods include mean filling, median filling, interpolation filling and regression filling;
[0018] The unification unit is used to standardize the relevant data information in the processing unit using the dimensionless technology, and convert the data of different dimensions into a dimensionless unified form.
[0019] Preferably, the early warning control module includes an early warning standard analysis unit and a judgment unit;
[0020] The warning standard analysis unit is used to process the gestational age Tz and the birth weight Ctz dimensionlessly according to the relevant personality characteristic data information, and map the corresponding data values to the interval In, construct the early warning degree factor Ycyz, and the early warning degree factor Ycyz is obtained by the following formula: ; In the formula, a and b are weight values, and e is the Euler number.
[0021] Preferably, the judgment unit is used to calculate the warning degree factors Ycyz of several newborns in the neonatal intensive care unit according to the method of obtaining the warning degree factor Ycyz in the warning standard analysis unit, and calculate the average warning degree factor Ycyz of several groups of warning degree factors Ycyz by statistics. and the standard deviation of the warning degree factor ; and based on the average warning severity factor and the standard deviation of the warning degree factor , set the No. 1 warning threshold through the following formula :
[0022] ; In the formula, k represents a constant;
[0023] A set of data columns is obtained by sorting several groups of warning degree factors Ycyz by numerical size, and the data columns are divided into three parts: front, middle and back, and different constant k values are set for the three parts, wherein the value of the constant k is 1-3, the value of the constant k of the front part is set to 3, the value of the constant k of the middle part is set to 2, and the value of the constant k of the back part is set to 1; wherein the sorting method is: from small to large order.
[0024] Preferably, the preliminary identification module includes an abnormality identification unit, an early warning unit and a historical statistics unit;
[0025] The abnormality identification unit is used to obtain the mean of each parameter in the relevant physiological state data information according to the relevant physiological state data information of the newborn in combination with the statistical mean algorithm, wherein the cardiac output Xsc is obtained by the following formula: Among them, SV represents the stroke volume, Pwtt represents the pulse wave conduction time;
[0026] The warning unit is used to calculate the average value of each parameter in the relevant physiological status data information and combine it with the first warning threshold , identify abnormal physiological parameters, the specific judgment contents are as follows:
[0027] If the mean value of the corresponding physiological parameters in the relevant physiological status data information falls within the first warning threshold When the newborn is detected to be in an abnormal state, no preliminary warning instructions will be issued.
[0028] If the mean value of the corresponding physiological parameters in the relevant physiological status data information does not fall within the No. 1 warning threshold When the newborn's physiological parameters are abnormal, it will be identified and marked as abnormal physiological parameters. At this time, a preliminary warning instruction will be issued to the outside.
[0029] The historical statistics unit is used to count the corresponding physiological parameters of the newborns in abnormal states under nursing intervention during the historical period according to the way of analyzing and identifying abnormal physiological parameters in the abnormal identification unit and the early warning unit, so as to generate a historical abnormality set.
[0030] Preferably, the abnormal analysis module is used to collect statistics on abnormal physiological parameters and obtain an abnormal index set after receiving the preliminary warning instruction. If the abnormal index set contains the same abnormal physiological parameters as those in the historical abnormal set, the abnormal physiological parameters of the newborn are preliminarily classified as non-pathological abnormalities.
[0031] At the same time, according to the abnormal indicator set, the abnormal parameter retention time Ydsc and the abnormal deviation value Ypcz are dimensionless, and the corresponding data values are mapped to the interval The abnormal evaluation coefficient Ygxs of the corresponding abnormal physiological parameters is constructed, which is obtained by the following formula: ; In the formula, is expressed as the number of exception parameters, , and are weight values, Expressed as the first correction constant.
[0032] Preferably, constructing the early warning score index Ypzs according to the abnormal indicator set includes:
[0033] According to the abnormal indicator set, the early warning score index Ypzs is obtained by the following formula: ; Where i=1, 2, 3, ..., n, n represents the number of abnormal physiological parameters in the abnormal indicator set, Expressed as the abnormal assessment coefficient of the ith abnormal physiological parameter, Expressed as the average anomaly assessment coefficient, Represents the breath sound of the ith abnormal physiological parameter, Expressed as the average breath sound, and are weight values, Expressed as the second correction constant.
[0034] Preferably, the scoring module includes a comparison unit and an intervention unit;
[0035] The comparison unit is used to obtain a first warning threshold according to the judgment unit Set the second warning threshold , by combining the warning score index Ypzs with the second warning threshold Comparative analysis is performed to comprehensively determine the current dangerous state of the newborn. The specific contents are as follows:
[0036] If the warning score index Ypzs falls into the second warning threshold When the abnormal physiological parameters of the newborn are classified as non-pathological abnormalities;
[0037] If the warning score index Ypzs does not fall within the second warning threshold When the abnormal physiological parameters of the newborn are classified as pathological abnormalities;
[0038] The intervention unit is used to take corresponding intervention measures according to the judgment content in the comparison unit. The specific intervention measures are:
[0039] When abnormal physiological parameters are classified as non-pathological abnormalities, the frequency of postural care is adjusted to 70% of the original frequency;
[0040] When abnormal physiological parameters are classified as pathological abnormalities, an alarm notification will be automatically sent to the medical team, reminding medical staff to manually check and evaluate the current condition of the newborn, and shorten the original monitoring interval to 70% of the original.
[0041] In a second aspect, the present invention provides a high-risk neonatal early warning scoring method, comprising:
[0042] Conduct real-time monitoring of newborns in the neonatal intensive care unit to obtain relevant personality characteristics data and relevant physiological status data information of the newborns;
[0043] Preprocess the relevant personality characteristics data information and relevant physiological status data information, remove noise, smooth the image, and perform geometric correction to generate a monitoring data set, and use dimensionless processing technology to standardize the relevant data information in the monitoring data set;
[0044] According to the relevant personality characteristic data information, the warning degree factor Ycyz is constructed, and the warning threshold is set according to the value of the warning degree factor Ycyz. ;
[0045] Based on the relevant physiological status data information of the newborn, combined with the No. 1 warning threshold , initially identify abnormal physiological parameters and issue preliminary warning instructions;
[0046] After receiving the preliminary warning instruction, the abnormal physiological parameters are counted to obtain the abnormal indicator set, and the warning score index Ypzs is constructed according to the abnormal indicator set;
[0047] An evaluation threshold is set in advance, and the evaluation threshold is compared and analyzed with the early warning score index Ypzs to comprehensively judge the current dangerous state of the newborn and take corresponding intervention measures according to the dangerous state.
[0048] The present invention provides a high-risk neonatal early warning scoring system and method, which has the following beneficial effects:
[0049] The status monitoring module monitors the newborns in the neonatal intensive care unit in real time. The system can continuously obtain the personality characteristics and physiological status data of the newborns. This real-time monitoring capability can promptly capture any abnormal physiological changes of the newborns, ensuring that an early warning signal can be issued immediately when an abnormal situation occurs, further avoiding the risks caused by delayed monitoring. The early warning control module constructs the early warning degree factor Ycyz and sets the No. 1 early warning threshold , so that the system can adaptively adjust the early warning standards according to the personalized characteristics and current physiological status of different newborns. This dynamic control mechanism not only improves the accuracy of the early warning, but also effectively avoids the false alarm or missed alarm caused by the fixed threshold. The combination of the preliminary identification module and the abnormal analysis module enables the system to not only preliminarily identify abnormal physiological parameters, but also generate an abnormal indicator set after further analysis, and construct an early warning scoring index Ypzs. Through the multi-dimensional analysis of abnormal physiological parameters, the system can more accurately judge the severity and potential risks of abnormal conditions, and provide a scientific basis for subsequent intervention measures. The scoring module can make a comprehensive judgment on the health status of newborns by setting the evaluation threshold and comparing it with the early warning scoring index Ypzs. According to the judgment results, the system can intelligently select appropriate intervention measures, so as to timely and effectively respond to the pathological abnormalities of high-risk newborns in actual nursing, reduce the risk of critical illness of newborns, and improve the quality and safety of nursing.
[0050] The early warning standard analysis unit performs dimensionless processing on gestational age Tz and birth weight Ctz and maps the data to a unified interval, effectively eliminating the impact of different newborn personality characteristics on the early warning factor. This processing method ensures that the calculation of the early warning degree factor Ycyz can accurately reflect the true health status of each newborn, thereby improving the personalization and pertinence of the early warning. The judgment unit calculates the average value and standard deviation of several groups of early warning degree factors Ycyz through statistical methods, and sets the No. 1 early warning threshold based on this. ,This method can dynamically adjust the threshold to adapt to the health changes of different newborn groups.,By setting different constants k, the system can set reasonable early warning standards for different health risk levels,according to the distribution of newborns, thereby optimizing the sensitivity and accuracy of the early warning.
[0051] The judgment unit sorts the warning degree factors and divides the data column into three parts: front, middle and back, and sets a different constant k for each part. This segmented processing can better distinguish the different degrees of health risks, ensure that the warning signal can accurately reflect the risk status of the newborn, and properly handle different levels of abnormalities. This method further avoids excessive or insufficient warnings.
[0052] The system can mark and record abnormal data after each nursing intervention in a timely manner, ensuring that the handling of each abnormal event is based on sufficient data analysis, thereby optimizing neonatal care and management decisions, improving nursing results and neonatal safety. Accurate mean calculation and threshold-based abnormal judgment reduce the risk of false positives and false negatives caused by data fluctuations. The comprehensive analysis capability of the historical statistical unit further reduces the misjudgment caused by a single abnormal event, ensuring the stability and reliability of the early warning system.
[0053] By calculating the early warning score index Ypzs and comparing it with the second early warning threshold After comparison, the system can further accurately classify abnormal physiological parameters as non-pathological or pathological. This classification not only improves the system's ability to identify different types of abnormalities, but also reduces the risk of false positives and negative positives. The identification of non-pathological abnormalities helps prevent excessive intervention, while the identification of pathological abnormalities ensures the timely handling of serious problems. When the system identifies non-pathological abnormalities, it can effectively reduce the fluctuations in physiological parameters caused by nursing intervention by adjusting the frequency of positional care to 70% of the original and optimizing the nursing process. For situations classified as pathological abnormalities, the system automatically sends an alarm to notify the medical team to ensure that medical staff can quickly learn of the abnormal state of the newborn. This timely alarm mechanism allows the medical team to immediately conduct manual inspections and evaluations to ensure that the pathological abnormalities of the newborn are promptly handled and intervened, thereby improving the safety and quality of care for the newborn. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] Figure 1 A block diagram of a high-risk neonatal early warning scoring system of the present invention;
[0055] Figure 2 The present invention is a flowchart of a high-risk neonatal early warning scoring method. DETAILED DESCRIPTION
[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0057] Embodiment 1 of the present invention
[0058] See also Figure 1 , the present invention provides a high-risk neonatal early warning scoring system, including a state monitoring module, a data preprocessing module, an early warning control module, a preliminary identification module, an abnormal analysis module and a scoring module;
[0059] The state monitoring module is used to monitor the newborns in the neonatal intensive care unit (NICU) in real time to obtain the relevant personality characteristics data information and relevant physiological state data information of the newborns;
[0060] The data preprocessing module is used to preprocess the relevant personality characteristic data information and the relevant physiological status data information, remove noise, smooth the image and perform geometric correction, generate a monitoring data set, and use dimensionless processing technology to standardize the relevant data information in the monitoring data set;
[0061] The warning control module is used to construct a warning degree factor Ycyz based on relevant personality characteristic data information, and set a first warning threshold according to the value of the warning degree factor Ycyz. ;
[0062] The preliminary identification module is used based on the relevant physiological status data information of the newborn and combined with the first warning threshold , identify abnormal physiological parameters and issue preliminary warning instructions;
[0063] The abnormal analysis module is used to perform statistics on abnormal physiological parameters after receiving the preliminary warning instruction to obtain an abnormal indicator set, and construct a warning score index Ypzs according to the abnormal indicator set;
[0064] The scoring module is used to pre-set an evaluation threshold, and compare and analyze the evaluation threshold with the early warning scoring index Ypzs to comprehensively determine the current dangerous state of the newborn and take corresponding intervention measures according to the dangerous state.
[0065] During the operation of this system, through the real-time monitoring of newborns in the intensive care unit (NICU) by the status monitoring module, the system can further continuously obtain the personality characteristics data and physiological status data of the newborns. This continuous monitoring ensures the timely grasp of the physiological condition of the newborn and further reduces the risks caused by delayed or missing monitoring. The data preprocessing module improves the quality of monitoring data through steps such as noise removal, image smoothing and geometric correction. The dimensionless processing technology standardizes the data to make the data more reliable and consistent, effectively reducing the risk of false alarms or missed reports caused by data quality issues. The early warning control module constructs the early warning degree factor Ycyz based on the personality characteristics data, and sets the No. 1 early warning threshold, thereby achieving a more accurate early warning threshold setting. Through this module, the system can better identify potential risk conditions, thereby improving the accuracy of early warnings. The preliminary identification module can , identify the abnormal physiological parameters of the newborn and issue preliminary warning instructions to ensure that a quick response can be made when an abnormality occurs, reducing the health risks caused by delays. After receiving the preliminary warning instructions, the abnormal analysis module generates a set of abnormal indicators and constructs a warning scoring index Ypzs by statistics on abnormal physiological parameters. This function enables the system to analyze the physiological abnormalities of the newborn from multiple dimensions and provide a more comprehensive risk assessment. The scoring module comprehensively judges the current dangerous state of the newborn by comparing the warning scoring index Ypzs with the preset assessment threshold, and takes corresponding intervention measures. This function ensures the timeliness and pertinence of the intervention measures, thereby reducing the health risks of the newborn as much as possible to improve the overall nursing effect. In summary, through the mutual cooperation of various modules, the system can realize real-time monitoring, accurate warning and timely intervention of high-risk newborns, greatly improving the quality and safety of newborn care.
[0066] Embodiment 2 of the present invention
[0067] Please refer to Figure 1 , specifically: the state monitoring module includes a No. 1 data acquisition unit and a No. 2 data acquisition unit;
[0068] The first data acquisition unit is used to monitor and record relevant personality characteristic data information of newborns in the neonatal intensive care unit in real time, wherein the relevant personality characteristic data information includes the birth weight Ctz and gestational age Tz of several newborns in the neonatal intensive care unit;
[0069] The No. 2 data acquisition unit is used to monitor and record relevant physiological status data information of the newborn in real time, wherein the relevant physiological status data information includes the heart rate parameters, blood pressure parameters, respiratory rate parameters, blood oxygen saturation parameters, body temperature parameters, blood sugar level and cardiac output Xsc of the newborn in each monitoring period under nursing intervention.
[0070] The data preprocessing module includes a processing unit and a unifying unit;
[0071] The processing unit is used to use a filter to identify and filter out noise in the relevant personality feature data information and the relevant physiological status data information to ensure the purity of the signal, and use a bilateral filtering method to smooth it, so as to remove subtle noise or defects in the image, reduce small fluctuations or artifacts in the image, and improve the clarity and analyzability of the image, thereby improving the accuracy of subsequent image analysis and feature extraction; and fill in missing values, wherein the methods of filling in missing values include mean filling, median filling, interpolation filling and regression filling;
[0072] The unified unit is used to standardize the relevant data information in the processing unit using dimensionless technology, converting data of different dimensions into a dimensionless unified form, thereby ensuring the comparability and consistency of various types of data in subsequent analysis.
[0073] In this embodiment, the state monitoring module can monitor and record the multi-dimensional data of the newborn in real time through the No. 1 data acquisition unit and the No. 2 data acquisition unit. The No. 1 data acquisition unit focuses on obtaining the individual characteristics of the newborn, while the No. 2 data acquisition unit obtains the physiological state data. This multi-dimensional data acquisition ensures that the system has a sufficient information basis when analyzing the health status of the newborn. The data preprocessing module further improves the purity and accuracy of the data through the noise identification and filtering, bilateral filtering and missing value filling of the processing unit. The filter removes the noise and minor defects in the data, and the bilateral filtering method reduces the artifacts and small fluctuations of the image, thereby improving the clarity of the image. These processing steps ensure the accuracy of subsequent analysis and feature extraction. The dimensionless technology of the unified unit converts data of different dimensions into a unified dimensionless form, ensuring the comparability and consistency of various types of data in subsequent analysis. This processing step makes the data have the same analysis basis under different dimensions and units, thereby further avoiding the analysis errors caused by data scale differences and improving the accuracy and reliability of the overall data processing. Through the careful processing and standardization of data, the data preprocessing module provides high-quality data support for subsequent anomaly detection, early warning and scoring analysis.
[0074] Embodiment 3 of the present invention
[0075] Please refer to Figure 1 ,Specifically: the early warning control module includes an early warning standard analysis unit and a judgment unit;
[0076] The warning standard analysis unit is used to process the gestational age Tz and the birth weight Ctz dimensionlessly according to the relevant personality characteristic data information, and map the corresponding data values to the interval In, construct the early warning degree factor Ycyz, and the early warning degree factor Ycyz is obtained by the following formula:
[0077] ; In the formula, a and b are weight values, and e is the Euler number, which is approximately 2.71828.
[0078] The above-mentioned gestational age Tz is usually recorded by the obstetrician at birth, based on the calculation of the gestational period, which is determined by the pregnant woman's gestational records and ultrasound examination.
[0079] Birth weight Ctz is accurately measured at birth using an electronic baby scale.
[0080] The judgment unit is used to calculate the warning degree factors Ycyz of several newborns in the neonatal intensive care unit according to the method of obtaining the warning degree factor Ycyz in the warning standard analysis unit, and calculate the average warning degree factor Ycyz of several groups of warning degree factors Ycyz through statistics. and the standard deviation of the warning degree factor ; and based on the average warning severity factor and the standard deviation of the warning degree factor , set the No. 1 warning threshold through the following formula :
[0081] ;
[0082] In the formula, k is a constant, usually with a value of 1-3, corresponding to different confidence levels;
[0083] A set of data columns is obtained by sorting several groups of warning degree factors Ycyz by numerical size, and the data columns are divided into three parts: front, middle and back, and different constant k values are set for the three parts, wherein the value of the constant k is 1-3, the value of the constant k of the front part is set to 3, the value of the constant k of the middle part is set to 2, and the value of the constant k of the back part is set to 1; wherein the sorting method is: from small to large order.
[0084] In this embodiment, the warning standard analysis unit performs dimensionless processing on the gestational age Tz and the birth weight Ctz, and maps the data values into a unified interval, thereby constructing a warning degree factor Ycyz. This processing method ensures that the data are comparable under different measurement scales, and enhances the accuracy and consistency of the warning. The judgment unit performs statistical analysis on the warning degree factors Ycyz of multiple newborns to calculate the average warning degree factor and its standard deviation. This method can reflect the distribution of early warning conditions in the newborn population, thus providing a basis for setting the No. 1 early warning threshold. Provides a scientific basis. By sorting and segmenting the data, combined with different constant k values (ranging from 1 to 3) to set the warning threshold, the setting of the warning threshold is more adaptable and flexible, and can be adjusted for different confidence levels, enhancing the accuracy and applicability of the system. The warning degree factor Ycyz is sorted and segmented, and the data column is divided into three parts: front, middle and back, and different constant values are set for each part. This multi-level risk classification method can divide the risk status of newborns in detail according to different warning degrees. The classification of high risk, medium risk and low risk enables the early warning system to provide corresponding response measures and intervention suggestions when facing different risk levels, so as to better manage the health status of newborns, and fully consider the individual differences of newborns and the fluctuations of group data when dealing with actual situations, enhancing the stability and adaptability of the system in different environments. Through statistical analysis and dynamic adjustment of the warning threshold method, high-risk newborns can be identified more accurately, reducing the risk of false alarms and missed reports. Accurate early warning response can take timely intervention measures when newborns are in potential danger, improving the safety and effectiveness of newborn care.
[0085] Embodiment 4 of the present invention
[0086] Please refer to Figure 1 ,Specifically: the preliminary identification module includes an abnormality identification unit, an early warning unit and a historical statistics unit;
[0087] The abnormality identification unit is used to obtain the mean of each parameter in the relevant physiological state data information according to the relevant physiological state data information of the newborn in combination with the statistical mean algorithm, wherein the cardiac output Xsc is obtained by the following formula: ; Among them, SV represents the stroke volume, and Pwtt represents the pulse wave conduction time;
[0088] The above-mentioned stroke volume SV is monitored and obtained by a cardiac ultrasound device. The heart's stroke volume can be measured by echocardiography, which can provide real-time images of the heart's structure and function and calculate and analyze the stroke volume.
[0089] The pulse wave conduction time Pwtt is monitored and obtained by a pulse wave conduction time monitor;
[0090] The warning unit is used to calculate the average value of each parameter in the relevant physiological status data information and combine it with the first warning threshold , identify abnormal physiological parameters, the specific judgment contents are as follows:
[0091] If the mean value of the corresponding physiological parameters in the relevant physiological status data information falls within the first warning threshold When the newborn is detected to be in an abnormal state, no preliminary warning instructions will be issued.
[0092] If the mean value of the corresponding physiological parameters in the relevant physiological status data information does not fall within the No. 1 warning threshold When the newborn's physiological parameters are abnormal, it will be identified and marked as abnormal physiological parameters. At this time, a preliminary warning instruction will be issued to the outside.
[0093] The historical statistics unit is used to count the corresponding physiological parameters of the newborns in abnormal states under nursing intervention during the historical period according to the way of analyzing and identifying abnormal physiological parameters in the abnormal identification unit and the early warning unit, so as to generate a historical abnormality set.
[0094] In this embodiment, the abnormality identification unit calculates the mean of each parameter in the relevant physiological status data information through a statistical mean algorithm, ensuring accurate assessment of the physiological status of the newborn, especially the calculation of cardiac output Xsc, which is obtained through the formula of stroke volume SV and pulse wave conduction time Pwtt, making the assessment of physiological parameters more accurate and better reflecting the real health status of the newborn. The early warning unit calculates the mean of each parameter and the set No. 1 early warning threshold , and accurately identify abnormalities. If the parameter mean falls within the threshold range, the physiological state is considered normal; if not, it is identified as abnormal. This judgment method based on a clear threshold can effectively screen out abnormal physiological parameters, further avoid errors caused by subjective judgment, and ensure the reliability of preliminary warnings. When abnormal physiological parameters are identified, the early warning unit can promptly issue preliminary warning instructions to the outside. This mechanism can ensure that when the physiological state of the newborn is abnormal, the nursing staff can quickly receive the alarm and intervene in time, thereby reducing potential health risks. The historical statistics unit generates a historical abnormality set by accumulating the analysis results of the abnormal identification unit and the early warning unit. This function enables the system to review and count previous abnormal situations, which helps to identify common abnormal patterns and trends, and provide data support for future early warning strategies, thereby improving the long-term prediction and adaptability of the system. By analyzing historical abnormal data, nursing staff can identify abnormal physiological parameter patterns that may occur under different nursing interventions, which provides a valuable basis for optimizing nursing intervention strategies, helps to formulate more effective nursing plans and adjust intervention measures, and improves the personalization and scientificity of neonatal care. Through statistics and analysis of historical data, the system can continuously optimize its recognition and early warning algorithms, and improve its adaptability to different health conditions of newborns.
[0095] Embodiment 5 of the present invention
[0096] Please refer to Figure 1Specifically: the abnormal analysis module is used to count the abnormal physiological parameters and obtain the abnormal index set after receiving the preliminary warning instruction. If the abnormal index set contains the same abnormal physiological parameters as those in the historical abnormal set, the abnormal physiological parameters of the newborn are preliminarily classified as non-pathological abnormalities;
[0097] At the same time, according to the abnormal indicator set, the abnormal parameter retention time Ydsc and the abnormal deviation value Ypcz are dimensionless, and the corresponding data values are mapped to the interval The abnormal evaluation coefficient Ygxs of the corresponding abnormal physiological parameters is constructed, which is obtained by the following formula:
[0098] ;
[0099] In the formula, is expressed as the number of exception parameters, , and are weight values, Expressed as the first correction constant, where 0 < ≤1,0< ≤1,0< ≤1, and + + =1.
[0100] The above abnormal parameter retention time Ydsc can be monitored and obtained through a vital sign monitor;
[0101] The abnormal deviation value Ypcz refers to the difference obtained by comparing the abnormal parameter with its mean;
[0102] Unusual number of parameters It is used for statistical acquisition.
[0103] In this embodiment, after receiving the preliminary warning instruction, the abnormal analysis module can further effectively distinguish non-pathological abnormalities from potential pathological abnormalities by counting abnormal physiological parameters and obtaining abnormal index sets. When the abnormal physiological parameters in the abnormal index set match the parameters in the historical abnormal set, the module can preliminarily classify these parameters as non-pathological abnormalities. This classification mechanism helps to reduce false alarms caused by nursing interventions or other non-pathological factors, and further ensures that only abnormalities that really need attention are further processed. By constructing the abnormal evaluation coefficient Ygxs, the impact of different abnormal parameters can be standardized, thereby ensuring the fairness and consistency of the evaluation. This accurate evaluation can help medical personnel better understand the severity and duration of abnormal states and improve the accurate judgment of the health status of newborns. Through accurate classification and standardized evaluation, the system can respond more accurately to each abnormal situation. Abnormal parameters can be further correctly classified as non-pathological abnormalities or abnormalities that require further attention, ensuring the accuracy of the early warning mechanism. The system can identify and exclude those abnormalities that do not require intervention at an early stage, and deal with real health risks in a timely manner. The abnormality analysis module can effectively use past data experience to judge current abnormalities by comparing with historical abnormality sets. This comparative analysis of historical data not only improves the system's ability to identify abnormal conditions of newborns, but also provides a valuable reference for future warning adjustments.
[0104] Embodiment 6 of the present invention
[0105] The early warning score index Ypzs is obtained by the following formula:
[0106] ;
[0107] Where i=1, 2, 3, ..., n, n represents the number of abnormal physiological parameters in the abnormal indicator set, Expressed as the abnormal assessment coefficient of the ith abnormal physiological parameter, Expressed as the average anomaly assessment coefficient, Represents the breath sound of the ith abnormal physiological parameter, Expressed as the average breath sound, and are weight values, Expressed as the second correction constant.
[0108] The abnormal physiological parameters mentioned above, such as breath sounds Hxy, can be monitored and obtained through an electronic stethoscope;
[0109] The scoring module includes a comparison unit and an intervention unit;
[0110] The comparison unit is used to obtain a first warning threshold according to the judgment unit Set the second warning threshold , by combining the warning score index Ypzs with the second warning threshold Comparative analysis is performed to comprehensively determine the current dangerous state of the newborn. The specific contents are as follows:
[0111] If the warning score index Ypzs falls into the second warning threshold When the abnormal physiological parameters of the newborn are classified as non-pathological abnormalities;
[0112] If the warning score index Ypzs does not fall within the second warning threshold When the abnormal physiological parameters of the newborn are classified as pathological abnormalities;
[0113] The intervention unit is used to take corresponding intervention measures according to the judgment content in the comparison unit. The specific intervention measures are:
[0114] When abnormal physiological parameters are classified as non-pathological abnormalities, the frequency of positional care is adjusted to 70% of the original frequency. By adjusting the nursing process, the fluctuation of physiological parameters caused by nursing intervention can be reduced, and the health status of the newborn can be better managed.
[0115] When abnormal physiological parameters are classified as pathological abnormalities, an alarm will be automatically sent to the medical team to remind medical staff to manually check and evaluate the current condition of the newborn, ensuring that the medical team can be informed of the newborn's pathological abnormalities at the first time and make necessary clinical judgments and interventions; and shorten the original monitoring interval to 70% of the original.
[0116] In this embodiment, the scoring module calculates the early warning score index Ypzs and compares it with the second early warning threshold By comparing, it is possible to further effectively distinguish whether the abnormal physiological parameters of the newborn are non-pathological or pathological. This precise classification method ensures that the system can take appropriate response measures for different abnormal conditions, thereby improving the reliability and practicality of the early warning system. When the abnormal physiological parameters are classified as non-pathological, the intervention unit will adjust the frequency of position care to 70% of the original. By adjusting the nursing process and reducing the fluctuation of physiological parameters caused by nursing intervention, this optimization measure can effectively reduce the false abnormalities caused by excessive care, thereby more accurately reflecting the true health status of the newborn and improving the scientificity and effectiveness of nursing. For situations classified as pathological abnormalities, the system will automatically send an alarm to notify the medical team, reminding medical staff to conduct manual examination and evaluation of the newborn. This timely alarm mechanism can ensure that the medical team is informed of the pathological abnormalities of the newborn at the first time, and quickly make necessary clinical judgments and interventions, thereby improving the emergency response speed and treatment efficiency. For pathological abnormalities, the monitoring interval of newborns will be shortened to 70% of the original. This measure ensures that the status of newborns can be monitored more frequently in the case of pathological abnormalities, so as to timely discover and respond to possible health problems and improve the management ability of the health status of newborns. Through clear early warning classification and targeted intervention measures, the system can optimize the allocation of medical resources, ensure that resources can be used first for pathological abnormalities that require high attention, reduce unnecessary interventions and examinations, and improve the utilization efficiency of medical resources.
[0117] Embodiment 7 of the present invention
[0118] Please refer to Figure 2 The present invention also provides a high-risk neonatal early warning scoring method, comprising:
[0119] Step 1: Conduct real-time monitoring of newborns in the neonatal intensive care unit to obtain relevant personality characteristics data and relevant physiological status data of the newborns;
[0120] Step 2: Preprocess the relevant personality characteristics data information and relevant physiological status data information, remove noise, smooth the image, and perform geometric correction to generate a monitoring data set, and use dimensionless processing technology to standardize the relevant data information in the monitoring data set;
[0121] Step 3: Based on the relevant personality characteristics data, construct the warning level factor Ycyz, and set the warning threshold according to the value of the warning level factor Ycyz ;
[0122] Step 4: Based on the relevant physiological status data of the newborn, combined with the No. 1 warning threshold , initially identify abnormal physiological parameters and issue preliminary warning instructions;
[0123] Step 5: After receiving the preliminary warning instruction, the abnormal physiological parameters are counted to obtain the abnormal indicator set, and the warning score index Ypzs is constructed according to the abnormal indicator set;
[0124] Step 6: Pre-set an assessment threshold, and compare and analyze the assessment threshold with the early warning score index Ypzs to comprehensively determine the current dangerous state of the newborn and take corresponding intervention measures according to the dangerous state.
[0125] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A high-risk newborn warning scoring system, characterized in that: include: Condition monitoring module, data preprocessing module, early warning control module, preliminary identification module, abnormal analysis module and scoring module; The state monitoring module is used to monitor the newborns in the neonatal intensive care unit in real time to obtain relevant personality characteristic data information and relevant physiological state data information of the newborns; The data preprocessing module is used to preprocess the relevant personality characteristic data information and the relevant physiological status data information, remove noise, smooth the image and perform geometric correction, generate a monitoring data set, and use dimensionless processing technology to standardize the relevant data information in the monitoring data set; The warning control module is used to construct a warning degree factor Ycyz based on relevant personality characteristic data information, and set a first warning threshold according to the value of the warning degree factor Ycyz. ; The early warning control module includes an early warning standard analysis unit and a judgment unit; The relevant personality characteristic data information includes the birth weight Ctz and gestational age Tz of several newborns in the neonatal intensive care unit; The warning standard analysis unit is used to process the gestational age Tz and the birth weight Ctz dimensionlessly according to the relevant personality characteristic data information, and map the corresponding data values to the interval In, construct the early warning degree factor Ycyz, and the early warning degree factor Ycyz is obtained by the following formula: ; In the formula, a and b are weight values, and e represents the Euler number; The judgment unit is used to calculate the warning degree factors Ycyz of several newborns in the neonatal intensive care unit according to the method of obtaining the warning degree factor Ycyz in the warning standard analysis unit, and calculate the average warning degree factor Ycyz of several groups of warning degree factors Ycyz through statistics. and the standard deviation of the warning degree factor ; Based on the average warning severity factor and the standard deviation of the warning degree factor , set the No. 1 warning threshold through the following formula ; ; In the formula, k represents a constant; By sorting the values of several groups of warning degree factors Ycyz, a set of data columns is obtained, and the data columns are divided into three parts: front, middle and back, and different constants k are set for the three parts, wherein the value of constant k is 1-3, the value of constant k of the front part is set to 3, the value of constant k of the middle part is set to 2, and the value of constant k of the back part is set to 1; wherein the sorting method is: from small to large order; The preliminary identification module is used based on the relevant physiological status data information of the newborn and combined with the first warning threshold , identify abnormal physiological parameters and issue preliminary warning instructions; The abnormal analysis module is used to perform statistics on abnormal physiological parameters after receiving the preliminary warning instruction to obtain an abnormal indicator set, and construct a warning score index Ypzs according to the abnormal indicator set; The abnormal analysis module is used to collect statistics on abnormal physiological parameters and obtain an abnormal index set after receiving the preliminary warning instruction. If the abnormal index set contains the same abnormal physiological parameters as those in the historical abnormal set, the abnormal physiological parameters of the newborn are preliminarily classified as non-pathological abnormalities. At the same time, according to the abnormal indicator set, the abnormal parameter retention time Ydsc and the abnormal deviation value Ypcz are dimensionless, and the corresponding data values are mapped to the interval The abnormal evaluation coefficient Ygxs of the corresponding abnormal physiological parameters is constructed, which is obtained by the following formula: ; In the formula, is expressed as the number of exception parameters, , and are weight values, Expressed as the first correction constant; The construction of the early warning score index Ypzs according to the abnormal indicator set includes: According to the abnormal indicator set, the early warning score index Ypzs is obtained by the following formula: ; Where i=1, 2, 3, ..., n, n represents the number of abnormal physiological parameters in the abnormal indicator set, Expressed as the abnormal assessment coefficient of the ith abnormal physiological parameter, Expressed as the average anomaly assessment coefficient, Represents the breath sound of the i-th abnormal physiological parameter, Expressed as the average breath sound, and are weight values, Expressed as the second correction constant; The scoring module is used to pre-set an evaluation threshold, and compare and analyze the evaluation threshold with the early warning scoring index Ypzs to comprehensively determine the current dangerous state of the newborn and take corresponding intervention measures according to the dangerous state.
2. The high-risk neonatal early warning scoring system according to claim 1, characterized in that: The state monitoring module includes a first data acquisition unit and a second data acquisition unit; The first data acquisition unit is used to monitor and record relevant personality characteristic data information of newborns in the neonatal intensive care unit in real time; The No. 2 data acquisition unit is used to monitor and record relevant physiological status data information of the newborn in real time, wherein the relevant physiological status data information includes the heart rate parameters, blood pressure parameters, respiratory rate parameters, blood oxygen saturation parameters, body temperature parameters, blood sugar level and cardiac output Xsc of the newborn in each monitoring period under nursing intervention.
3. The high-risk neonatal early warning scoring system according to claim 1, characterized in that: The data preprocessing module includes a processing unit and a unifying unit; The processing unit is used to identify and filter out noise in the relevant personality characteristic data information and the relevant physiological status data information by using a filter, and smooth them by using a bilateral filtering method, and fill in missing values, wherein the missing value filling methods include mean filling, median filling, interpolation filling and regression filling; The unification unit is used to standardize the relevant data information in the processing unit using the dimensionless technology, and convert the data of different dimensions into a dimensionless unified form.
4. The high-risk neonatal early warning scoring system according to claim 2, characterized in that: The preliminary identification module includes an abnormality identification unit, an early warning unit and a historical statistics unit; The abnormality identification unit is used to obtain the mean of each parameter in the relevant physiological state data information according to the relevant physiological state data information of the newborn in combination with the statistical mean algorithm, wherein the cardiac output Xsc is obtained by the following formula: ; Among them, SV represents the stroke volume, and Pwtt represents the pulse wave conduction time; The warning unit is used to calculate the average value of each parameter in the relevant physiological status data information and combine it with the first warning threshold , identify abnormal physiological parameters, the specific judgment contents are as follows: If the mean value of the corresponding physiological parameters in the relevant physiological status data information falls within the No. 1 warning threshold When the newborn is detected to be in an abnormal state, no preliminary warning instructions will be issued. If the mean value of the corresponding physiological parameters in the relevant physiological status data information does not fall within the No. 1 warning threshold When the newborn's physiological parameters are abnormal, it will be identified and marked as abnormal physiological parameters. At this time, a preliminary warning instruction will be issued to the outside. The historical statistics unit is used to count the corresponding physiological parameters of the newborns in abnormal states under nursing intervention during the historical period according to the way of analyzing and identifying abnormal physiological parameters in the abnormal identification unit and the early warning unit, so as to generate a historical abnormality set.
5. The high-risk neonatal early warning scoring system according to claim 1, characterized in that: The scoring module includes a comparison unit and an intervention unit; The comparison unit is used to obtain the first warning threshold according to the judgment unit Set the second warning threshold , by combining the warning score index Ypzs with the second warning threshold Comparative analysis is performed to comprehensively determine the current dangerous state of the newborn. The specific contents are as follows: If the warning score index Ypzs falls into the second warning threshold When the abnormal physiological parameters of the newborn are classified as non-pathological abnormalities; If the warning score index Ypzs does not fall within the second warning threshold When the abnormal physiological parameters of the newborn are classified as pathological abnormalities; The intervention unit is used to take corresponding intervention measures according to the judgment content in the comparison unit. The specific intervention measures are: When abnormal physiological parameters are classified as non-pathological abnormalities, the frequency of postural care is adjusted to 70% of the original frequency; When abnormal physiological parameters are classified as pathological abnormalities, an alarm notification will be automatically sent to the medical team, reminding medical staff to manually check and evaluate the current condition of the newborn, and shorten the original monitoring interval to 70% of the original.
6. A high-risk neonatal early warning scoring method, used to implement the high-risk neonatal early warning scoring system according to any one of claims 1 to 5, characterized in that: include: Conduct real-time monitoring of newborns in the neonatal intensive care unit to obtain relevant personality characteristics and physiological status data of the newborns; Preprocess the relevant personality characteristics data information and relevant physiological status data information, remove noise, smooth the image, and perform geometric correction to generate a monitoring data set, and use dimensionless processing technology to standardize the relevant data information in the monitoring data set; According to the relevant personality characteristic data information, the warning degree factor Ycyz is constructed, and the warning threshold is set according to the value of the warning degree factor Ycyz. ; Based on the relevant physiological status data information of the newborn, combined with the No. 1 warning threshold , initially identify abnormal physiological parameters and issue preliminary warning instructions; After receiving the preliminary warning instruction, the abnormal physiological parameters are counted to obtain the abnormal indicator set, and the warning score index Ypzs is constructed according to the abnormal indicator set; An evaluation threshold is set in advance, and the evaluation threshold is compared and analyzed with the early warning score index Ypzs to comprehensively judge the current dangerous state of the newborn and take corresponding intervention measures according to the dangerous state.
