Intelligent monitoring and analyzing system for physiological data of anesthetized patient

Abstract

The application provides a physiological data intelligent monitoring and analysis system for anesthetized patients, and belongs to the field of data processing.The method comprises the following steps: acquiring physiological data sequences of an anesthetized patient in several dimensions to obtain several intervals of each dimension; obtaining noise data according to the difference of different physiological data in each interval of each dimension, removing the noise data in each interval of each dimension to obtain each normal interval of each dimension; obtaining a correction value of each physiological data according to the distribution of physiological data in several monotonic intervals in each normal interval of each dimension, and further obtaining a predicted peak value in the next interval of the interval in which the current time is located in each dimension. The application aims to solve the problem that the current prediction algorithm does not consider the influence of noise and ignores the slight change of data when predicting data, so that the reference of the prediction result is not strong.

Description

Technical Field

[0001] This invention belongs to the field of data processing, specifically relating to an intelligent monitoring and analysis system for physiological data of anesthetized patients. Background Technology

[0002] Monitoring the physiological data of anesthetized patients is a core aspect of ensuring surgical safety. Currently, electrocardiogram (ECG) monitors are used to capture cardiac electrical activity, and non-invasive cuff blood pressure monitors or invasive arterial catheters (real-time blood pressure waveforms) are used to assess circulatory status. Central venous pressure (CVP) monitoring is used to assess cardiac preload and volume status.

[0003] Predicting the physiological data of anesthetized patients is one of the core development directions of modern intelligent anesthesia. Its fundamental purpose can be summarized as: shifting from passive monitoring to active prediction and intervention, thereby maximizing patient safety, optimizing the quality of anesthesia, and improving long-term prognosis, achieving a leap from "passive alarm" to "active defense".

[0004] When collecting patients' physiological data, noise is inevitably generated due to machine vibration or sensor data transmission, causing drastic fluctuations in the data at any given moment, resulting in a significant discrepancy from the patient's normal data. Current traditional early warning algorithms lack noise filtering capabilities, leading to significant impacts on subsequent predictions due to the large difference between the generated noise and normal data. Furthermore, current prediction algorithms exhibit a certain lag in data processing, failing to promptly predict subsequent data risks and prone to false alarms. Consequently, the predicted values ​​obtained using current algorithms cannot effectively assist doctors in treatment. Summary of the Invention

[0005] To address the issues that traditional early warning algorithms fail to remove noise from data and ignore minute fluctuations in the collected data, resulting in predictions that do not effectively assist doctors in treatment when using current early warning algorithms to predict the physiological data of anesthetized patients, this invention proposes an intelligent monitoring and analysis system for the physiological data of anesthetized patients.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a data acquisition module: acquiring physiological data sequences of anesthetized patients in several dimensions; a data analysis module: dividing the physiological data sequences of anesthetized patients in each dimension into several intervals of equal size, obtaining several intervals in each dimension; obtaining the noise performance of each physiological data based on the difference between each physiological data in each interval of each dimension and the mean of all physiological data in the same interval, thereby obtaining noise data; removing noise data in each interval, obtaining each normal interval in each dimension; using the extreme points in each normal interval of each dimension as segmentation points, obtaining several monotonic intervals in each normal interval of each dimension; and based on each normal interval of each dimension... Physiological data at both ends and physiological data at both ends of each normal interval are used to obtain the degree of change in each normal interval and the degree of change in each monotonic interval of each normal interval for each dimension. Combined with each physiological data point in each monotonic interval of each normal interval for each dimension, a correction value for each physiological data point is obtained. Data prediction module: Based on the distribution of correction values ​​of physiological data in the last few normal intervals of each dimension and the degree of change in the last few normal intervals of each dimension, the predicted peak value of physiological data in each dimension in the next interval of the current time interval is obtained to assist doctors in treating patients; the current time interval is the last normal interval of each dimension.

[0007] Furthermore, the specific steps for obtaining several intervals for each dimension are as follows: using a data acquisition time of... The window is divided into segments, and based on the collection time corresponding to each physiological data point, the anesthetized patient is divided into segments at the specified time. The physiological data sequence of each dimension is divided into several windows; among them, The preset collection duration; the anesthetized patient will be on the [number]th [day / month]. Within the physiological data sequence of the first dimension The window, denoted as the 1st window. The first dimension The interval; and the first interval; The first dimension The interval and the first Each interval does not contain physiological data collected at the same time.

[0008] Furthermore, the specific steps for obtaining the noise performance of each physiological data point are as follows: based on the first... The first dimension The values ​​of each physiological data point within each interval and its comparison with other physiological data are used to obtain the first... The first dimension The variability of each physiological data point within each interval; according to the... The first dimension The variability of each physiological data point within a given interval is compared with the variability of all physiological data points within the interval containing that physiological data point to obtain the ... The first dimension Noise performance of each physiological data point within each interval; obtain the noise performance of the first interval. The first dimension Within the interval, the first The specific formula for calculating the variability of physiological data is as follows: In the formula, Indicates the first The first dimension Within the interval, the first The variability of individual physiological data Indicates the first The first dimension Within the interval, the first The numerical values ​​of each physiological data point Indicates the first The first dimension The mean of all physiological data within each interval This represents the absolute value function.

[0009] Furthermore, the obtained first The first dimension The specific formula for calculating the noise performance of each physiological data point within each interval is as follows: In the formula, Indicates the first The first dimension Within the interval, the first Noise representation of physiological data Indicates the first The first dimension Within the interval, the first The variability of individual physiological data Indicates the first The first dimension The mean of the variability of all physiological data within a given interval Indicates the first The first dimension The minimum value among all physiological data fluctuations within a given interval. Represents the absolute value function. express function.

[0010] Furthermore, the specific steps for obtaining the noise data are as follows: Preset a noise performance threshold. ,like Then the first The first dimension Within the interval, the first Each physiological data point is denoted as noise data.

[0011] Furthermore, the specific calculation formulas for obtaining the degree of change in each normal interval of each dimension and the degree of change in each monotonic interval within each normal interval are as follows: In the formula, Indicates the first The first dimension The degree of change within the normal range Indicates the first The first dimension The value of the first physiological data within the normal range. Indicates the first The first dimension The value of the last physiological data within the normal range; In the formula, Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Indicates the first The first dimension The first of the normal intervals The value of the first physiological data within a monotonic interval. Indicates the first The first dimension The first of the normal intervals The value of the last physiological data within a monotonic interval.

[0012] Furthermore, the specific steps for obtaining the correction value for each physiological data point are as follows: based on the first... The first dimension The degree of change of all monotonic intervals in the normal intervals is obtained to determine the th... The first dimension The degree of data variation within each monotonic interval of a normal interval; according to the... The first dimension The degree of change of each monotonic interval in the normal interval is related to the first... The first dimension The degree of change in the first normal interval is obtained to determine the first... The first dimension The trends of each normal interval and each of its monotonic intervals are consistent; according to the... The first dimension The magnitude of data variation within each monotonic interval of the normal interval, the first... The first dimension The consistency of trends between the normal intervals and each of the monotonic intervals within them, the first... The first dimension The degree of change of each normal interval and each monotonic interval therein, and the... The first dimension The value of each physiological data point within each monotonic interval of each normal interval is obtained. The first dimension The first of the normal intervals The correction value for each physiological data point within each monotonic interval; obtain the correction value for the first... The first dimension The specific formula for calculating the drastic change in data within each monotonic interval of a normal interval is as follows: In the formula, Indicates the first The first dimension The first of the normal intervals The degree of drastic change in data within a monotonic interval Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Indicates the first The first dimension The mean of the absolute values ​​of the degree of change of all monotonic intervals in a normal interval. Indicates the first The first dimension The minimum absolute value of the degree of change among all monotonic intervals in a normal interval. This represents Iverson brackets. If the input satisfies the condition in the brackets, output 1; otherwise, output 0.

[0013] Furthermore, the obtained first The first dimension The specific formula for calculating the trend consistency between each normal interval and each monotonic interval within it is as follows: In the formula, Indicates the first The first dimension The normal interval and the first one in it The trend consistency of each monotonic interval Indicates the first The first dimension The degree of change within the normal range Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Represents a symbolic function. Represents the absolute value function. This represents Iverson brackets. If the input satisfies the condition in the brackets, output 1; otherwise, output 0.

[0014] Furthermore, the obtained first The first dimension The first of the normal intervals The specific formula for calculating the correction value of each physiological data point within a monotonic interval is as follows: In the formula, Indicates the first The first dimension The first of the normal intervals Within the _ monotonic intervals, the _th Correction values ​​for each physiological data point Indicates the first The first dimension The first of the normal intervals Within the _ monotonic intervals, the _th The numerical values ​​of each physiological data point Indicates the first The first dimension The degree of change within the normal range Indicates the first The first dimension The normal interval and the first one in it The trend consistency of each monotonic interval Indicates the first The first dimension The first of the normal intervals The degree of drastic change in data within a monotonic interval Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Indicates the first The first dimension The minimum absolute value of the degree of change among all monotonic intervals in a normal interval. Indicates the first The first dimension The maximum value among the absolute values ​​of the degree of change of all monotonic intervals in a normal interval. Represents the absolute value function. Represents a symbolic function.

[0015] Furthermore, the specific calculation formula for the predicted peak value of each dimension of physiological data in the next interval of the current time interval is as follows: In the formula, Indicates the first The predicted peak value of each dimension of physiological data in the next interval of the current time interval. Indicates the first The correction value of the last physiological data in the last normal interval of each dimension. Indicates the first The sum of the degree of change in the last three normal intervals of each dimension. Represents an ordinal value. Indicates the first The reciprocal of the first dimension The maximum value among all corrected physiological data within a normal range. Indicates the first The reciprocal of the first dimension The minimum value among all corrected physiological data within a normal range. Indicates the first The reciprocal of the first dimension The mean of the corrected values ​​of all physiological data within a normal range This represents the weight normalization function.

[0016] The intelligent monitoring and analysis system for physiological data of anesthetized patients provided by this invention has the following beneficial effects: When determining whether each physiological data point is abnormal, this invention first divides the physiological data sequence into several intervals based on the data collection time. Then, it calculates the noise performance of each physiological data point by dividing the difference between the mean of all physiological data points within the same interval and the difference between the mean of all physiological data points within that interval and the smallest physiological data point within that interval. This solves the problem that if only the difference between each physiological data point and other physiological data points is used to quantify the noise performance of each physiological data point, it may lead to issues such as the gradual decrease in anesthetic drug content in the patient's body over time, resulting in different normal fluctuation ranges for patients in different time intervals. This could potentially lead to some normal data being misclassified as noise data and vice versa. Furthermore, when predicting data in the next time interval after the current time based on data from the current time and data before the current time, this invention assigns greater weight to data in time intervals closer to the current time, making the prediction results more accurate. Furthermore, when predicting data peaks in the next time interval after the current time, this invention addresses the issue that patients' metabolic rates are low during anesthesia, which may lead to relatively low fluctuations in related physiological data. These fluctuations are often minor and easily overlooked by the machine over short periods. However, over a long period, these physiological data changes can accumulate to a significant extent. This invention divides the data within each time interval into multiple monotonic intervals. Then, it adjusts the values ​​of physiological data within the monotonic intervals with more dramatic fluctuations and in the same direction of change as the data in its respective time interval. This allows the current method to capture minute changes and improves the reliability of the predicted peaks. Attached Figure Description

[0017] To more clearly illustrate the embodiments and design schemes of the present invention, the accompanying drawings required for this embodiment will be briefly described below. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of an intelligent monitoring and analysis system for physiological data of anesthetized patients according to an embodiment of the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand and implement the technical solutions of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0020] Example 1: This invention provides an intelligent monitoring and analysis system for physiological data of anesthetized patients, specifically as follows: Figure 1 As shown, it includes: Step S001: Obtain physiological data sequences of anesthetized patients in several dimensions.

[0021] Specifically, a patient undergoing anesthesia during surgery is designated as an anesthetized patient. Physiological data collected during anesthesia is retrieved from the hospital's system to obtain the patient's physiological data. A sequence of physiological data in several dimensions. Each physiological data point corresponds to a specific acquisition time. The preset number of dimensions in this embodiment... The selected dimensions are blood pressure, systolic blood pressure dynamic fluctuation, cardiac output, and stroke volume. The preset number of dimension data and the selected dimensions can be set to other values ​​in other implementations.

[0022] Thus, we obtained a sequence of physiological data for the anesthetized patient in several dimensions.

[0023] Step S002: Divide the physiological data sequence of the anesthetized patient in each dimension into several intervals of the same size to obtain several intervals in each dimension; based on the difference between each physiological data in each interval of each dimension and the mean of all physiological data in the same interval, obtain the noise performance of each physiological data, and thus obtain the noise data; remove the noise data in each interval to obtain each normal interval in each dimension.

[0024] It should be noted that when collecting physiological data such as dynamic systolic blood pressure (SBP), heart rate, and blood pressure of anesthetized patients through relevant machines, some noise is inevitably generated due to machine vibration or data transmission by sensors. This can cause drastic fluctuations in the relevant data at a certain moment, resulting in a significant difference from the patient's normal data. Such data can also interfere with the subsequent analysis of the anesthetized patient's physiological data, leading to errors in subsequent judgments. Therefore, noisy data in the physiological data sequence is removed.

[0025] It should be further explained that, due to the significant difference between noisy data and surrounding normal data, noise data is obtained based on the difference between each physiological data point and its surrounding physiological data. Because the anesthetic drugs are gradually metabolized during surgery, the normal fluctuation range of the patient's physiological data at different time points constantly changes, making it impossible to directly process data from all moments to obtain noisy data. Therefore, the real-time data sequence for each dimension is divided into multiple small intervals. The volatility of each data point within each interval is obtained based on the overall difference between each data point and all surrounding data. Then, the noise performance of each physiological data point is calculated based on the difference in volatility between each data point within an interval and all surrounding data.

[0026] It should be further noted that as the surgery progresses, the anesthetic drugs in the patient's body are continuously metabolized, causing the normal fluctuation range of physiological data in the same dimension within different intervals to continuously increase. Therefore, the normal fluctuation range of all data within each interval is quantified by the difference between the variability of all physiological data within that interval and the minimum variability of the physiological data within that interval. Combining the difference in overall variability between each data point within an interval and all surrounding data, the noise performance of each physiological data point is obtained.

[0027] Specifically, the data collection time is The window is divided into segments, and based on the collection time corresponding to each physiological data point, the anesthetized patient is divided into segments at the specified time. The physiological data sequence of each dimension is divided into several windows. The anesthetized patient is then... Within the physiological data sequence of the first dimension The window, denoted as the 1st window. The first dimension There are several intervals. The use of a window to divide the sequence is a well-known technique and will not be detailed in this embodiment. Furthermore, different windows do not contain physiological data of the same dimension collected at the same time, i.e., the [missing information - likely a specific interval or range]. The first dimension The interval and the first Each interval does not contain physiological data collected at the same time. The preset collection duration in this embodiment... This example is used to illustrate the concept; other values ​​can be set in other implementations.

[0028] Furthermore, obtain the first The first dimension Within the interval, the first The specific formula for calculating the variability of physiological data is as follows: In the formula, Indicates the first The first dimension Within the interval, the first The variability of individual physiological data Indicates the first The first dimension Within the interval, the first The numerical values ​​of each physiological data point Indicates the first The first dimension The mean of all physiological data within each interval This represents the absolute value function.

[0029] It should be noted that, The higher the value, the more likely the anesthetized patient was in the first... The first dimension Within the interval, the first The greater the difference between a physiological data point and other physiological data points within the same interval, the more it conforms to the characteristic that noisy data differs significantly from its surrounding normal data. The first dimension Within the interval, the first The greater the likelihood that a physiological data point is noise.

[0030] Furthermore, obtain the first The first dimension Within the interval, the first The specific formula for calculating the noise performance of each physiological data point is as follows: In the formula, Indicates the first The first dimension Within the interval, the first Noise representation of physiological data Indicates the first The first dimension Within the interval, the first The variability of individual physiological data Indicates the first The first dimension The mean of the variability of all physiological data within a given interval Indicates the first The first dimension The minimum value among all physiological data fluctuations within a given interval. Represents the absolute value function. express The function, in this embodiment, is used for normalization processing.

[0031] It should be noted that, The smaller the value, the better compared to the first... The first dimension Other physiological data within each interval, the first The first dimension Within the interval, the first The fluctuation of the first physiological data is relatively small, that is, the first The first dimension Within the interval, the first The physiological data is likely to be normal. It is used to reflect the first The first dimension The normal fluctuation range of physiological data within each interval; the smaller the value, the better. The first dimension The fluctuation of normal physiological data within a given interval is relatively small; at this time, if If the value is large, it indicates that the first... The first dimension Within the interval, the first The likelihood that these physiological data are normal is relatively low; The larger the value, the more significant the first... The first dimension The fluctuations in normal physiological data within a given interval are significant; if at this time... If the value is small, it means that the first The first dimension Within the interval, the first The physiological data is likely to be normal. The larger the value, the more significant the first... The first dimension Within the interval, the first The greater the likelihood that a physiological data point is noise data.

[0032] Furthermore, if Then the first The first dimension Within the interval, the first The first physiological data point is denoted as noise data. Remove the first... The first dimension The noise data within the interval is denoted as the nth interval. The first dimension A normal range. Among them, the preset noise performance threshold in this embodiment. This example is used to illustrate the concept; other values ​​can be set in other implementations.

[0033] Thus, we have obtained each normal interval for each dimension.

[0034] Step S003: Take the extreme points in each normal interval of each dimension as segmentation points to obtain several monotonic intervals in each normal interval of each dimension; based on the physiological data of the left and right ends of each normal interval of each dimension and the physiological data of the left and right ends of each monotonic interval in each normal interval, obtain the degree of change of each normal interval of each dimension and the degree of change of each monotonic interval in each normal interval; combine each physiological data in each monotonic interval of each normal interval of each dimension to obtain the correction value of each physiological data.

[0035] It should be noted that when predicting the data value for the next acquisition time based on data collected at the current time and other acquisition times prior to the current time, the fluctuations in the relevant physiological data of anesthetized patients are relatively low due to the reduced basal metabolic rate and decreased impact of endogenous stimulation on circulation during anesthesia, especially in short periods where changes are minimal and easily overlooked by the machine. However, over a long period, these physiological data changes accumulate to a significant degree, at which point human intervention may not be effective. Therefore, to determine whether the patient's data has undergone relatively drastic changes in a short period, it is necessary to enhance some data to create a greater difference compared to normal data, ensuring that the computer does not ignore these data in subsequent operations. Thus, adjustments are made to the values ​​of some data within the physiological data sequence.

[0036] It's important to further clarify that when a patient's physiological data in one dimension gradually changes in one direction, the patient's physiological data within the normal range in that dimension also changes in the same direction. However, this directional change in physiological data occurs through continuous fluctuations. For example, if a patient's blood pressure increases, over a longer timeframe, there's a trend of increasing blood pressure; over a shorter timeframe, blood pressure may fluctuate, but the increasing trend is stronger than the decreasing trend. Therefore, the normal range for each dimension is divided into several monotonic intervals, and the amplitude of change in each monotonic interval is obtained.

[0037] It should be further noted that when an anesthetized patient's physiological data shows an upward trend within a normal range in one dimension, to prevent this trend from being overlooked due to its small magnitude, the upward-growing interval within the normal range should be adjusted. This adjustment should specifically target the monotonic interval with a larger upward increase, as this better reflects the upward trend.

[0038] It should be further noted that when anesthetized patients exhibit a downward trend in physiological data within a normal range, adjustments should be made to the downward-growing intervals within the normal range to prevent this trend from being overlooked due to its small magnitude. Specifically, adjustments should be made to the monotonic intervals with a larger downward decrease. Lowering the monotonic intervals with a larger downward decrease better reflects the downward trend within a normal range. Therefore, adjustments should be made to the data within the monotonic intervals that exhibit the same direction of change as the normal range.

[0039] It should be further explained that, within a normal interval, the difference between the mean of the variation amplitudes of all monotonic intervals and the minimum variation amplitude of the monotonic interval can effectively reflect the normal variation amplitude within that interval. Therefore, by subtracting the difference between the variation amplitude and the minimum variation amplitude of each monotonic interval from the variation amplitude of that monotonic interval, and comparing this difference with the difference between the mean of the variation amplitudes of all monotonic intervals and the minimum variation amplitude of the monotonic interval, we can identify monotonic intervals with drastic changes. Then, we adjust the data within these monotonic intervals that meet the criteria for data adjustment.

[0040] Specifically, to obtain the first The first dimension The maximum and minimum points within a normal interval. Obtaining the maximum and minimum points within a data set is a well-known technique and will not be elaborated upon in this embodiment.

[0041] Furthermore, the first The first dimension Using the maximum and minimum points within the normal intervals as segmentation points, the first... The first dimension The normal interval is divided into several monotonic intervals. Among them, the extreme points are the maximum and minimum points.

[0042] Furthermore, obtain the first The first dimension The specific formula for calculating the degree of change in each normal interval is as follows: In the formula, Indicates the first The first dimension The degree of change within the normal range Indicates the first The first dimension The value of the first physiological data within the normal range. Indicates the first The first dimension The value of the last physiological data within the normal range.

[0043] It should be noted that, This indicates that the anesthetized patient was in the first... The first dimension The normal range shows an upward trend. This indicates that the anesthetized patient was in the first... The first dimension The normal range shows a downward trend; This indicates that it is not necessary to perform anesthesia on the first day. The first dimension The physiological data within the normal range were adjusted.

[0044] Furthermore, obtain the first The first dimension The first of the normal intervals The specific formula for calculating the degree of change in each monotonic interval is as follows: In the formula, Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Indicates the first The first dimension The first of the normal intervals The value of the first physiological data within a monotonic interval. Indicates the first The first dimension The first of the normal intervals The value of the last physiological data within a monotonic interval.

[0045] It should be noted that, This indicates that the anesthetized patient was in the first... The first dimension The first of the normal intervals Each monotonic interval shows an upward trend. This indicates that the anesthetized patient was in the first... The first dimension The first of the normal intervals Each monotonic interval shows a downward decreasing trend; This indicates that it is not necessary to perform anesthesia on the first day. The first dimension The first of the normal intervals The physiological data within a monotonic interval were adjusted.

[0046] Furthermore, obtain the first The first dimension The first of the normal intervals The specific formula for calculating the drastic change in data within a monotonic interval is as follows: In the formula, Indicates the first The first dimension The first of the normal intervals The degree of drastic change in data within a monotonic interval Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Indicates the first The first dimension The mean of the absolute values ​​of the degree of change of all monotonic intervals in a normal interval. Indicates the first The first dimension The minimum absolute value of the degree of change among all monotonic intervals in a normal interval. This represents Iverson brackets. If the input satisfies the condition in the brackets, output 1; otherwise, output 0.

[0047] It should be noted that when At that time, it indicates that the anesthetized patient was in the first... The first dimension The first of the normal intervals If the data within a monotonic interval shows a drastic change, it may be necessary to re-evaluate the anesthetized patient on the [number]th [time / period]. The first dimension The first of the normal intervals Adjust the data within each monotonic interval; At that time, it indicates that the anesthetized patient was in the first... The first dimension The first of the normal intervals The data changes relatively smoothly within the monotonic intervals, so it is not necessary to monitor the anesthetized patient during the [number]th [interval]. The first dimension The first of the normal intervals Adjust the data within each monotonic interval.

[0048] Furthermore, obtain the first The first dimension The normal interval and the first one in it The specific formula for calculating the trend consistency of a monotonic interval is as follows: In the formula, Indicates the first The first dimension The normal interval and the first one in it The trend consistency of each monotonic interval Indicates the first The first dimension The degree of change within the normal range Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Represents a symbolic function. Represents the absolute value function. This represents Iverson brackets. If the input satisfies the condition in the brackets, output 1; otherwise, output 0.

[0049] It should be noted that when At that time, it indicates that the anesthetized patient was in the first... The first dimension The normal interval and the first one in it The data within each monotonic interval showed a consistent trend, which may be indicative of the anesthetized patient's condition during the first monotonic interval. The first dimension The first of the normal intervals Adjust the data within each monotonic interval, at this time... ;when When the time is 0 or 0, it indicates that the anesthetized patient is in the first day. The first dimension The normal interval and the first one in it The data trends within the monotonic intervals are inconsistent, and the anesthetized patient is not observed in the [number]th [interval]. The first dimension The first of the normal intervals Adjust the data within each monotonic interval.

[0050] Furthermore, obtain the first The first dimension The first of the normal intervals Within the _ monotonic intervals, the _th The specific formula for calculating the correction value of each physiological data point is as follows: In the formula, Indicates the first The first dimension The first of the normal intervals Within the _ monotonic intervals, the _th Correction values ​​for each physiological data point Indicates the first The first dimension The first of the normal intervals Within the _ monotonic intervals, the _th The numerical values ​​of each physiological data point Indicates the first The first dimension The degree of change within the normal range Indicates the first The first dimension The normal interval and the first one in it The trend consistency of each monotonic interval Indicates the first The first dimension The first of the normal intervals The degree of drastic change in data within a monotonic interval Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Indicates the first The first dimension The minimum absolute value of the degree of change among all monotonic intervals in a normal interval. Indicates the first The first dimension The maximum value among the absolute values ​​of the degree of change of all monotonic intervals in a normal interval. Represents the absolute value function. This function represents the sign function, which outputs 1 when the input is greater than 0, 0 when the input is equal to 0, and -1 when the input is less than 0.

[0051] It should be noted that when and When both are 1, it means that the first The first dimension The first of the normal intervals If the data within a monotonic interval meets the adjustment criteria, then the data within that monotonic interval needs to be adjusted; when and When one of them is not 1 or all of them are not 1, then No, not the first The first dimension The first of the normal intervals Adjust the values ​​of the data within each monotonic interval; when When, explain the first The first dimension If a normal interval shows an upward trend, and multiple monotonic intervals within that interval are adjusted, then the data within the monotonic intervals that meet the adjustment criteria will be increased; when When, explain the first The first dimension If a normal interval has a downward decreasing trend, and multiple monotonic intervals within this interval are adjusted, then the values ​​of the data in the monotonic intervals within this interval that meet the data adjustment criteria will be reduced. The larger the value, the more significant the first... The first dimension The first of the normal intervals If the data within a monotonic interval varies significantly, then if... If so, then a large adjustment will be made to the data within that monotonic interval.

[0052] Thus, the correction value for each physiological data point in each normal interval of each dimension is obtained.

[0053] Step S004: Based on the distribution of the corrected values ​​of physiological data in the last few normal intervals of each dimension, and the degree of change in the last few normal intervals of each dimension, obtain the predicted peak value of the physiological data of each dimension in the next interval of the current time interval, to assist doctors in treating patients.

[0054] It's important to note that when predicting data after the current time by analyzing the distribution of correction values ​​within the normal interval of the current time and several normal intervals preceding it, the earlier the data points in the current time interval and preceding intervals have a smaller impact on the subsequent prediction results. Therefore, when predicting the peak data within the next interval after the current time interval, different weights are assigned to the data in different intervals.

[0055] It's important to further clarify that because the physiological data of anesthetized patients are constantly changing, predicting subsequent data requires using the last data point within the current time interval as the basis for prediction. Furthermore, for different data trends, such as an upward trend, the predicted subsequent data should also exhibit the same trend. Additionally, the maximum potential range of fluctuation in subsequent data should be predicted based on the maximum range of fluctuation in previous data. This ensures that the patient's data remains relatively stable during anesthesia.

[0056] It should be further explained that, since the forecasting process requires considering the data trends across all normal time intervals, and earlier data has little impact on subsequent forecasts, only the data from the last few normal time intervals are used as a reference. Furthermore, the forecast is based on the data value at the last moment of the current time interval, and the peak value of the next interval is obtained by calculating the average fluctuation ratio of the data from the previous three normal time intervals.

[0057] It should be further noted that when calculating the average volatility ratio for each interval, this method uses the difference between the maximum or minimum correction value within an interval and the average correction value within that interval as the average volatility value. Then, the ratio of the average volatility value within each interval to the difference between the maximum and minimum values ​​within that interval is used as the average volatility ratio for that interval.

[0058] Specifically, to obtain the first The specific formula for calculating the predicted peak value of each dimension of physiological data in the next interval of the current time interval is as follows: In the formula, Indicates the first The predicted peak value of each dimension of physiological data in the next interval of the current time interval. Indicates the first The correction value of the last physiological data in the last normal interval of each dimension. Indicates the first The sum of the degree of change in the last three normal intervals of each dimension. Represents an ordinal value. Indicates the first The reciprocal of the first dimension The maximum value among all corrected physiological data within a normal range. Indicates the first The reciprocal of the first dimension The minimum value among all corrected physiological data within a normal range. Indicates the first The reciprocal of the first dimension The mean of the corrected values ​​of all physiological data within a normal range This represents the weight normalization function.

[0059] It should be noted that, , indicating the first The physiological data in the first dimension shows an upward trend, thus affecting the second dimension. The correction value of the physiological data with the longest acquisition time in the last normal interval of each dimension is adjusted upwards to obtain the 1st dimension. The predicted peak value of physiological data in each dimension within the next interval of the current time interval; , indicating the first The physiological data in the first dimension shows a downward trend, thus affecting the second dimension. The correction value of the physiological data with the longest acquisition time in the last normal interval of each dimension is adjusted downwards to obtain the _th _ ... The predicted peak value of physiological data in each dimension within the next interval of the current time interval; through Assign a larger weight to normal intervals that are closer to the current time; The value represents the first The reciprocal of the first dimension The fluctuation range of the maximum value relative to the average value within each interval, through Quantitative First The reciprocal of the first dimension The larger this value, the greater the average fluctuation ratio of the intervals. The reciprocal of the first dimension The greater the fluctuation ratio of the maximum value relative to the average value within a given interval, the more... At that time, To achieve a large increase; through Quantitative First The reciprocal of the first dimension The larger this value, the greater the average fluctuation ratio of the intervals. The reciprocal of the first dimension The greater the fluctuation ratio of the maximum value relative to the average value within a given interval, the more... At that time, Make a large reduction.

[0060] Furthermore, through the first The predicted peak values ​​of physiological data in each dimension within the next interval of the current time interval remind anesthesiologists and doctors to carefully examine the relevant data and assist doctors in analyzing the physiological data of anesthetized patients.

[0061] This concludes the invention.

Claims

1. A smart monitoring and analysis system for physiological data of anesthetized patients, characterized in that, include: Data acquisition module: Acquires physiological data sequences of anesthetized patients in several dimensions; Data Analysis Module: Divides the physiological data sequence of anesthetized patients in each dimension into several equally sized intervals, resulting in several intervals for each dimension; Based on the difference between each physiological data point in each interval of each dimension and the mean of all physiological data points in the same interval, the noise performance of each physiological data point is obtained, thus yielding noisy data; Noisy data is removed from each interval, resulting in each normal interval of each dimension; Extreme points within each normal interval of each dimension are used as segmentation points, resulting in several monotonic intervals within each normal interval of each dimension; Based on the physiological data at the left and right endpoints of each normal interval of each dimension and the physiological data at the left and right endpoints of each monotonic interval within each normal interval,... The system obtains the degree of change in each normal interval and the degree of change in each monotonic interval within each normal interval for each dimension. Combining this with each physiological data point within each monotonic interval of each normal interval for each dimension, a correction value for each physiological data point is obtained. The data prediction module, based on the distribution of correction values ​​for physiological data in the last few normal intervals of each dimension and the degree of change in the last few normal intervals of each dimension, obtains the predicted peak value of the physiological data for each dimension in the next interval after the current time interval, assisting doctors in treating patients. The current time interval is the last normal interval for each dimension. The specific calculation formula for the noise performance of each physiological data point is as follows: In the formula, Indicates the first The first dimension Within the interval, the first Noise representation of physiological data Indicates the first The first dimension Within the interval, the first The variability of individual physiological data Indicates the first The first dimension The mean of the variability of all physiological data within a given interval Indicates the first The first dimension The minimum value among all physiological data fluctuations within a given interval. Represents the absolute value function. express The specific calculation formula for obtaining the correction value of each physiological data point is as follows: In the formula, Indicates the first The first dimension The first of the normal intervals Within the _ monotonic intervals, the _th Correction values ​​for each physiological data point Indicates the first The first dimension The first of the normal intervals Within the _ monotonic intervals, the _th The numerical values ​​of each physiological data point Indicates the first The first dimension The degree of change within the normal range Indicates the first The first dimension The normal interval and the first one in it The trend consistency of each monotonic interval Indicates the first The first dimension The first of the normal intervals The degree of drastic change in data within a monotonic interval Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Indicates the first The first dimension The minimum absolute value of the degree of change among all monotonic intervals in a normal interval. Indicates the first The first dimension The maximum value among the absolute values ​​of the degree of change of all monotonic intervals in a normal interval. Represents the absolute value function. Represents a symbolic function.

2. The intelligent monitoring and analysis system for physiological data of anesthetized patients according to claim 1, characterized in that, The specific steps for obtaining several intervals for each dimension are as follows: using a data acquisition time of... The window is divided into segments, and based on the collection time corresponding to each physiological data point, the anesthetized patient is divided into segments at the specified time. The physiological data sequence of each dimension is divided into several windows; among them, The preset collection duration; the anesthetized patient will be on the [number]th [day / month]. Within the physiological data sequence of the first dimension The window, denoted as the 1st window. The first dimension The interval; and the first interval; The first dimension The interval and the first Each interval does not contain physiological data collected at the same time.

3. The intelligent monitoring and analysis system for physiological data of anesthetized patients according to claim 1, characterized in that, Get the first The first dimension Within the interval, the first The specific formula for calculating the variability of physiological data is as follows: In the formula, Indicates the first The first dimension Within the interval, the first The variability of individual physiological data Indicates the first The first dimension Within the interval, the first The numerical values ​​of each physiological data point Indicates the first The first dimension The mean of all physiological data within each interval This represents the absolute value function.

4. The intelligent monitoring and analysis system for physiological data of anesthetized patients according to claim 1, characterized in that, The specific steps for obtaining noise data are as follows: Preset a noise performance threshold. ,like Then the first The first dimension Within the interval, the first Each physiological data point is denoted as noise data.

5. The intelligent monitoring and analysis system for physiological data of anesthetized patients according to claim 1, characterized in that, The specific calculation formulas for obtaining the degree of change of each normal interval for each dimension and the degree of change of each monotonic interval in each normal interval are as follows: In the formula, Indicates the first The first dimension The degree of change within the normal range Indicates the first The first dimension The value of the first physiological data within the normal range. Indicates the first The first dimension The value of the last physiological data within the normal range; In the formula, Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Indicates the first The first dimension The first of the normal intervals The value of the first physiological data within a monotonic interval. Indicates the first The first dimension The first of the normal intervals The value of the last physiological data within a monotonic interval.

6. The intelligent monitoring and analysis system for physiological data of anesthetized patients according to claim 1, characterized in that, Get the first The first dimension The first of the normal intervals The specific formula for calculating the drastic change in data within a monotonic interval is as follows: In the formula, Indicates the first The first dimension The first of the normal intervals The degree of drastic change in data within a monotonic interval Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Indicates the first The first dimension The mean of the absolute values ​​of the degree of change of all monotonic intervals in a normal interval. Indicates the first The first dimension The minimum absolute value of the degree of change among all monotonic intervals in a normal interval. This represents Iverson brackets. If the input satisfies the condition in the brackets, output 1; otherwise, output 0.

7. The intelligent monitoring and analysis system for physiological data of anesthetized patients according to claim 1, characterized in that, Get the first The first dimension The normal interval and the first one in it The specific formula for calculating the trend consistency of a monotonic interval is as follows: In the formula, Indicates the first The first dimension The normal interval and the first one in it The trend consistency of each monotonic interval Indicates the first The first dimension The degree of change within the normal range Indicates the first The first dimension The first of the normal intervals The degree of change in each monotonic interval Represents a symbolic function. Represents the absolute value function. This represents Iverson brackets. If the input satisfies the condition in the brackets, output 1; otherwise, output 0.

8. The intelligent monitoring and analysis system for physiological data of anesthetized patients according to claim 1, characterized in that, The specific calculation formula for the predicted peak value of each dimension of physiological data in the next interval of the current time interval is as follows: In the formula, Indicates the first The predicted peak value of each dimension of physiological data in the next interval of the current time interval. Indicates the first The correction value of the last physiological data in the last normal interval of each dimension. Indicates the first The sum of the degree of change in the last three normal intervals of each dimension. Represents an ordinal value. Indicates the first The reciprocal of the first dimension The maximum value among all corrected physiological data within a normal range. Indicates the first The reciprocal of the first dimension The minimum value among all corrected physiological data within a normal range. Indicates the first The reciprocal of the first dimension The mean of the corrected values ​​of all physiological data within a normal range This represents the weight normalization function.

Images