CN-121971046-A - Intelligent pregnancy venous thrombosis monitoring method and system

CN121971046ACN 121971046 ACN121971046 ACN 121971046ACN-121971046-A

Abstract

The invention relates to the technical field of venous thrombosis data monitoring, in particular to an intelligent pregnancy venous thrombosis monitoring method and system, comprising the following steps: and acquiring venous blood flow data during pregnancy, acquiring dynamic indexes of blood flow speed, blood pressure and blood viscosity, and judging whether the dynamic indexes fluctuate within a set steady-state range by setting the steady-state range to generate characteristic data of pregnancy health monitoring. According to the invention, indexes such as blood flow speed, blood pressure and blood viscosity during pregnancy are dynamically monitored, so that the change trend of the venous blood flow state can be efficiently identified, and the high-risk blood flow state can be identified in advance and intervened in time according to weight analysis of the transition frequency and duration time among the blood flow states in the state transition process. In addition, by analyzing the blood flow state transition trend, the deviation direction and the fluctuation amplitude of the high risk state can be more intuitively identified. The feature analysis is more specific by screening blood flow dynamic features closely related to the high risk state and eliminating irrelevant redundant information.

Inventors

ZHANG JUNRONG
ZHANG HUIWEN
ZHANG YUQUAN
CHENG XI
Tao Xixi

Assignees

南通大学附属医院

Dates

Publication Date: 20260505
Application Date: 20260129

Claims (8)

1. An intelligent pregnancy venous thrombosis monitoring method is characterized by comprising the following steps: Acquiring venous blood flow data during pregnancy, acquiring dynamic indexes of blood flow speed, blood pressure and blood viscosity, and judging whether the dynamic indexes fluctuate within a set steady-state range by setting the steady-state range to generate characteristic data of pregnancy health monitoring; According to the dividing result of blood flow data in the characteristic data of pregnancy health monitoring, counting the change condition of venous blood flow states in the stages, obtaining the transition weight of the states among the stages, constructing a blood flow state transition matrix by comparing the occurrence frequency and duration of the states with a preset risk threshold, and analyzing the deviation direction and amplitude of the high-risk states to obtain blood flow state transition characteristics; based on the data of the blood flow state transfer matrix in the blood flow state transfer characteristics, screening blood flow speed, blood pressure and blood viscosity dynamic index characteristics, classifying risk categories according to the screened dynamic index characteristics, and integrating pregnant women with similar transfer characteristics into the same group to generate a thrombus risk characteristic data set; And dynamically adjusting the distribution range parameter of the current feature based on the thrombus risk feature data set, determining a risk interval of the blood flow state by referring to the distribution range of the current feature and the historical deviation frequency, and generating a dynamic risk early warning signal.
2. The intelligent pregnancy venous thrombosis monitoring method according to claim 1, wherein the step of acquiring the characteristic data of pregnancy health monitoring specifically comprises the following steps: Converting the signal data of the time domain into a frequency domain based on the acquired pregnancy venous blood flow data, screening and inhibiting the interference signals, restoring the signals to the time domain, and carrying out subsection statistics to obtain a numerical distribution result of the venous blood flow data; according to the numerical distribution result of the venous blood flow data, adopting the formula: ; Calculating standard value Obtaining steady-state range reference data; Wherein, the Is the number of samples to be taken, Is the first of the sampled data A secondary measurement; and acquiring dynamic index data according to a set period, comparing the acquired data with the steady-state range reference data, and judging whether the blood flow speed, the blood pressure and the blood viscosity fluctuate within a reasonable range or not to obtain characteristic data of pregnancy health monitoring.
3. The method for intelligently monitoring venous thrombosis during pregnancy according to claim 2, wherein the step of obtaining the transition weight of the state between stages is specifically as follows: According to the characteristic data of pregnancy health monitoring, the change condition of venous blood flow state in the statistical stage adopts the formula: ; Calculating slave states in monitoring phase Transition to State Is the relative probability of occurrence of (a) Obtaining transition probability data between states; Wherein, the Is the slave state in the monitoring phase Transition to State Is used for the number of times of (a), Is the average duration between the states, Is the total number of states that are present, Representing slave states Transition to all other possible states Average duration of (2); and according to the transition probability data among the states, adopting a formula: ; Computing state To state Weight value of (2) Obtaining transition weights among states; Wherein, the Is a state of Transition to State Is used for the frequency of (a), Representing slave states Transition to all possible states Is a relative frequency of (a) is a relative frequency of (b).
4. The method for intelligently monitoring venous thrombosis during pregnancy according to claim 3, wherein the step of obtaining the blood flow state transition matrix is specifically as follows: Based on the occurrence frequency and duration of the monitoring data acquisition state, comparing the occurrence frequency and duration with a preset risk threshold, marking the state with the frequency or duration exceeding the threshold as a high-risk state, increasing the transition weight among the states of the high-risk state, and constructing a blood flow state transition matrix; And extracting a transfer path of the high-risk state according to the blood flow state transfer matrix, recording the transfer direction and frequency of the high-risk state, and analyzing the offset trend and direction of the high-risk state to obtain the blood flow state transfer characteristics.
5. The method for intelligent monitoring of venous thrombosis during pregnancy according to claim 4, wherein the step of obtaining risk classification according to the screened dynamic index features comprises the following steps: Extracting information of a high-risk state based on the data of the blood flow state transition matrix in the blood flow state transition characteristics, and evaluating the association degree between the high-risk state and the high-risk state according to the dynamic index characteristics to obtain a risk association analysis result; And according to the risk correlation analysis result, carrying out risk category classification on the dynamic index features, and carrying out dynamic change trend analysis on each category to obtain a grouped feature set.
6. The method for intelligently monitoring venous thrombosis during pregnancy according to claim 5, wherein the step of acquiring the thrombus risk characteristic data set is specifically as follows: And according to the grouped feature set, adopting a formula: ; Calculate the first Pregnant woman and the first Similarity value of individual pregnant women in same risk category Obtaining similarity data of dynamic index features; Wherein, the Representing a sequence of features in the current risk category, Is the total number of features contained within the current risk category, And Respectively represent the first Pregnant women and the first The pregnant woman is at the first The original data value under the individual characteristics is, Is the first The average value of the individual features is used, Is the first Standard deviation of individual features; and integrating pregnant women with similarity lower than a preset similarity threshold value into the same group according to the similarity data of the dynamic index features, and recording common high risk features of the pregnant women in each group to obtain a thrombus risk feature data set.
7. The method for intelligently monitoring venous thrombosis during pregnancy according to claim 6, wherein the step of acquiring the dynamic risk early warning signal is specifically as follows: Based on the thrombus risk characteristic data set, comparing blood flow velocity, blood pressure and blood viscosity data acquired in real time, recording deviation amplitude and frequency exceeding characteristic values in a steady-state range, dynamically updating the distribution range of blood flow velocity, blood pressure and blood viscosity characteristics, and generating an updated characteristic distribution range; And according to the updated characteristic distribution range, adopting a formula: ; Computing a given feature distribution In the case of (a), the current state is a risk state Posterior probability of (2) Generating a dynamic risk early warning signal by determining a risk interval of a blood flow state; Wherein, the Is expressed in a risk state Features are observed below Is a function of the probability of (1), Is the prior probability of being a priori, Is the feature probability.
8. An intelligent monitoring system for venous thrombosis during pregnancy is characterized in that, the method for intelligently monitoring venous thrombosis during pregnancy according to any one of claims 1 to 7, wherein the system comprises: The blood flow data acquisition module is used for detecting the stability of data and analyzing dynamic changes based on the pregnancy vein blood flow dynamic data by setting a stable numerical range, counting the fluctuation characteristics in a steady state, judging whether the data accords with the steady state range, and generating characteristic data for pregnancy health monitoring; The blood flow state analysis module counts the data change of venous blood flow in various time phases based on the characteristic data of pregnancy health monitoring, analyzes the occurrence frequency of the state of each phase, and identifies the fluctuation trend of the high risk state to obtain the blood flow state transfer characteristic; the dynamic feature screening module analyzes blood flow state transition matrix data based on the blood flow state transition features, determines indexes of blood flow speed, blood pressure and blood viscosity, and generalizes similar individuals of the blood flow states through data comparison of frequency and offset direction to generate a thrombus risk feature data set; The risk category grouping module divides the deviation frequency and the occurrence time of the feature data based on the thrombus risk feature data set, analyzes the risk level of the feature group and generates risk feature grouping data; And the risk early warning module analyzes the distribution interval of the real-time characteristic data based on the risk characteristic grouping data, combines the stepwise offset distribution, and confirms the risk area through the offset characteristic to generate a dynamic risk early warning signal.

Description

Intelligent pregnancy venous thrombosis monitoring method and system Technical Field The invention relates to the technical field of venous thrombosis data monitoring, in particular to an intelligent pregnancy venous thrombosis monitoring method and system. Background The venous thrombosis data monitoring technology tracks and evaluates the venous thrombosis risk in a patient in real time through means of medical imaging, blood analysis, data monitoring and the like. The field includes prevention, detection and management of thrombotic diseases such as deep vein thrombosis and pulmonary embolism to reduce the risk of thrombosis, improve diagnostic accuracy, and optimize therapeutic protocols. Wherein, pregnancy venous thrombosis intelligent monitoring is the venous thrombosis monitoring system that is used for pregnancy women. During pregnancy, women are at a significantly increased risk of venous thrombosis due to changes in hormone levels and changes in blood flow. The intelligent monitoring method is mainly used for monitoring the blood flow state, the blood coagulation index and the like of the pregnant woman in real time through wearable equipment or portable monitoring equipment in combination with data analysis and machine learning technology, so that potential thrombosis risks are rapidly identified, and health of the mother and the infant is guaranteed. The existing pregnancy venous thrombosis monitoring is difficult to effectively cope with the rapid change of blood flow state in terms of data processing, and the identification of the pregnant woman thrombus risk state is delayed. In addition, the monitoring of the blood flow state is difficult to analyze the blood flow characteristic differences at different stages, and the judgment of thrombus risk is easy to be inaccurate. And simultaneously, the potential high risk blood flow state is difficult to identify in time and intervene dynamically. In addition, the risk is inferred only through the staged data in the prior art, so that a risk early warning signal is difficult to timely and effectively send out, and the thrombus risk is predicted. Disclosure of Invention The invention aims to solve the defects in the prior art, and provides an intelligent pregnancy venous thrombosis monitoring method and system. In order to achieve the purpose, the invention adopts the following technical scheme that the intelligent pregnancy venous thrombosis monitoring method comprises the following steps: S1, acquiring venous blood flow data during pregnancy, acquiring dynamic indexes of blood flow speed, blood pressure and blood viscosity, and judging whether the dynamic indexes fluctuate within a set steady-state range by setting the steady-state range to generate characteristic data for pregnancy health monitoring; s2, counting the change condition of venous blood flow states in the stage according to the dividing result of blood flow data in the characteristic data of pregnancy health monitoring, obtaining the transition weight of the states among the stages, constructing a blood flow state transition matrix by comparing the occurrence frequency and duration time of the states with a preset risk threshold value, and analyzing the deviation direction and amplitude of a high-risk state to obtain blood flow state transition characteristics; S3, screening blood flow velocity, blood pressure and blood viscosity dynamic index features based on the data of the blood flow state transfer matrix in the blood flow state transfer features, classifying risk categories according to the screened dynamic index features, and integrating pregnant women with similar transfer features into the same group to generate a thrombus risk feature data set; and S4, dynamically adjusting the distribution range parameter of the current feature based on the thrombus risk feature data set, determining a risk interval of the blood flow state by referring to the distribution range and the historical deviation frequency of the current feature, and generating a dynamic risk early warning signal. As a further aspect of the present invention, the step of obtaining the characteristic data of pregnancy health monitoring specifically includes: s111, converting the signal data of the time domain into the frequency domain based on the acquired pregnancy venous blood flow data, screening and inhibiting the interference signals, restoring the signals to the time domain, and carrying out subsection statistics to obtain the numerical distribution result of the venous blood flow data; S112, according to the numerical distribution result of the venous blood flow data, adopting the formula: Calculating standard value Obtaining steady-state range reference data; Wherein, the Is the number of samples to be taken,Is the first of the sampled dataA secondary measurement; S113, collecting dynamic index data according to a set period, comparing the data collected each time with the steady-state range reference data,