CN-121971049-A - Real-time heart failure patient vital sign monitoring system and method

CN121971049ACN 121971049 ACN121971049 ACN 121971049ACN-121971049-A

Abstract

The application provides a real-time heart failure patient vital sign monitoring method and system, which are characterized by extracting corresponding physiological variability indexes from real-time monitoring data of each vital sign, constructing an individual physiological baseline model, generating a dynamic threshold interval of each vital sign, performing cross-parameter correlation analysis on deviation of the real-time monitoring data of each vital sign and the dynamic threshold interval to obtain pathology correlation indexes among abnormal changes of different vital signs, performing abnormal judgment on the real-time monitoring data of each vital sign, performing cross-validation on an abnormal judgment result of each vital sign to obtain an abnormal judgment confidence coefficient of each vital sign, and generating vital sign abnormal early warning information adapting to individual differences of a heart failure patient when the abnormal judgment confidence coefficient of each vital sign meets preset confidence conditions. By adopting the scheme of the application, the accurate adaptation of the vital sign static threshold to the individual difference of heart failure patients can be realized.

Inventors

LEI JIAJIA
FENG ZHIJIE
CAI QUANFENG

Assignees

阳江市人民医院

Dates

Publication Date: 20260505
Application Date: 20251224

Claims (10)

1. The heart failure risk early warning method based on multi-parameter fusion is characterized by comprising the following steps of: Acquiring real-time monitoring data of vital signs of a heart failure patient in a rehabilitation process; extracting corresponding physiological variability indexes from the real-time monitoring data of each vital sign, constructing an individual physiological baseline model according to all the physiological variability indexes and basic disease information of heart failure patients, and further generating a dynamic threshold interval of each vital sign; Performing cross-parameter association analysis on the deviation degree of the real-time monitoring data of each vital sign and the dynamic threshold interval based on a preset pathology association rule base to obtain pathology association indexes among different vital sign abnormal changes; Performing abnormality judgment on the real-time monitoring data of each vital sign according to all dynamic threshold intervals, and performing cross verification on an abnormality judgment result of each vital sign through the pathology correlation index to obtain an abnormality judgment confidence coefficient of each vital sign; and generating vital sign abnormality early warning information adapting to individual differences of heart failure patients when the abnormality judgment confidence degree of the vital signs meets a preset confidence condition.
2. The method of claim 1, wherein extracting the corresponding physiological variability index from the real-time monitoring data of each vital sign specifically comprises: Extracting time domain variation characteristics of real-time monitoring data of each vital sign; and determining the physiological variability index of each vital sign according to all the time domain variation characteristics.
3. The method of claim 1, wherein constructing an individualized physiological baseline model from all physiological variability indices and underlying disease information of heart failure patients specifically comprises: Acquiring basic disease information of heart failure patients, wherein the basic disease information comprises age, gender, duration of course, complications type and past hospitalization records; Constructing an initial physiological baseline model based on underlying disease information of heart failure patients; And carrying out iterative optimization on the initial physiological baseline model according to all physiological variability indexes to obtain an individualized physiological baseline model.
4. The method according to claim 1, wherein generating a dynamic threshold interval for each vital sign comprises: based on the individual physiological baseline model, extracting baseline reference values of all vital signs; determining a physiological fluctuation range of each vital sign around a baseline reference value; carrying out boundary correction on all physiological fluctuation ranges by combining the safety ranges of all vital signs in heart failure clinical diagnosis and treatment guidelines; and determining a dynamic threshold interval of each vital sign according to the baseline reference value of each vital sign and all the corrected physiological fluctuation ranges.
5. The method of claim 1, wherein performing cross-parameter correlation analysis on the deviation degree of the real-time monitoring data of each vital sign from the dynamic threshold interval based on a preset pathology correlation rule base, and obtaining pathology correlation indexes between abnormal changes of different vital signs specifically comprises: Calculating the deviation degree of the real-time monitoring data of each vital sign and the corresponding dynamic threshold interval, wherein the deviation degree is the ratio of the amplitude of the real-time monitoring data exceeding the upper limit and the lower limit of the threshold value to the width of the threshold interval; Invoking a preset pathology association rule base, wherein the pathology association rule base comprises association relations of all vital signs in a heart failure pathology mechanism; screening a plurality of vital sign parameter pairs with pathological relevance according to the pathological relevance rule base and all deviation degrees; For each vital sign parameter pair, calculating the synchronous change rate between two vital sign parameters in the vital sign parameter pair, wherein the synchronous change rate is the proportion of the time length, which is beyond a threshold interval, of the two vital sign parameters in the vital sign parameter pair to the total monitoring time length; and determining pathological relevance indexes among abnormal changes of different vital signs according to the importance degree of each vital sign in pathological mechanism and all synchronous change rates.
6. The method according to claim 1, wherein the performing abnormality determination on the real-time monitoring data of each vital sign according to all dynamic threshold intervals specifically comprises: Comparing the real-time monitoring data of each vital sign with the corresponding dynamic threshold interval one by one; If the monitoring data point in the real-time monitoring data is positioned in the dynamic threshold interval, judging that the monitoring data point is normal; If the monitoring data point in the real-time monitoring data exceeds the upper limit or the lower limit of the dynamic threshold interval, marking the monitoring data point as abnormal; counting the number of continuous abnormal monitoring data points, and judging that the vital sign is continuously abnormal when the number of the continuous abnormal monitoring data points reaches a preset number threshold; the level of abnormality and duration at which each vital sign is continuously abnormal are recorded.
7. The method of claim 1, wherein cross-validating the abnormality determination result for each vital sign by the pathology-relevance index, the obtaining the abnormality determination confidence for each vital sign specifically comprises: for each vital sign, acquiring a pathology correlation index of other vital signs with which pathology is associated; determining an abnormal association rate of vital signs according to the number of abnormal monitoring points which are also judged to be abnormal in the real-time monitoring data of the associated vital signs; determining an association verification value of the vital sign according to the abnormal association rate and the pathology association indexes of the vital sign and other vital signs with pathology association; determining a basic confidence value of the vital sign based on the duration of the vital sign abnormality determination and the abnormality level; and determining the abnormality judgment confidence of the vital sign according to the association verification value and the basic confidence value.
8. The heart failure patient vital sign real-time monitoring system comprises a heart failure risk early-warning unit, and is characterized in that the heart failure risk early-warning unit specifically comprises: the acquisition module is used for acquiring real-time monitoring data of vital signs of a heart failure patient in the rehabilitation process; the processing module is used for extracting corresponding physiological variability indexes from the real-time monitoring data of each vital sign, constructing an individual physiological baseline model according to all the physiological variability indexes and basic disease information of heart failure patients, and further generating dynamic threshold intervals of each vital sign; the processing module is further used for performing cross-parameter association analysis on the deviation degree of the real-time monitoring data of each vital sign and the dynamic threshold interval based on a preset pathology association rule base to obtain pathology association indexes among different vital sign abnormal changes; The processing module is further used for carrying out anomaly judgment on the real-time monitoring data of each vital sign according to all dynamic threshold intervals, and carrying out cross validation on the anomaly judgment result of each vital sign through the pathology correlation index to obtain the anomaly judgment confidence coefficient of each vital sign; and the execution module is used for generating vital sign abnormality early warning information adapting to individual differences of heart failure patients when the abnormality judgment confidence degree of the vital signs meets a preset confidence condition.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the heart failure patient vital sign real-time monitoring method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the method for real-time monitoring of vital signs of a heart failure patient according to any one of claims 1 to 7.

Description

Real-time heart failure patient vital sign monitoring system and method Technical Field The application relates to the technical field of vital sign monitoring, in particular to a method and a system for real-time monitoring of vital signs of heart failure patients. Background The prior vital sign monitoring technology mainly collects key physiological parameters of a human body through various sensors, including an electrocardio sensor, a photoelectric volume pulse wave sensor, a blood pressure detection module, a respiration monitoring device, a blood oxygen probe and the like, so as to obtain core indexes such as heart rate, blood pressure, respiratory rate, blood oxygen saturation and the like, the collected signals can be transmitted to a remote server or a medical management platform in a wireless communication mode after being preprocessed by a local processing unit, and based on cloud computing and an artificial intelligent algorithm, dynamic analysis, anomaly detection and trend prediction can be carried out on vital sign data, and continuous monitoring from a hospital to a home scene is realized. Vital sign monitoring is considered one of the central means in the long-term management of heart failure. Clinically, indexes such as heart rate, blood pressure, respiratory rate, blood oxygen saturation and the like are usually relied on for assessing and early warning the health state. However, the existing vital sign determination is mostly based on a unified static threshold setting, the static mode is difficult to fully reflect individual differences among different patients, heart failure patients are affected by age, disease progress, basic diseases, life style and other factors, the physiological tolerance ranges of the heart failure patients are significantly different, for example, part of patients can still maintain a stable state at a lower blood pressure level, and another part of patients can rapidly deteriorate even if slightly exceeds a conventional threshold, if the patients depend on a fixed threshold, false alarm or missing alarm is easy to occur, and the effectiveness of early warning and the feasibility of personalized management are limited, so that how to realize the accurate adaptation of the vital sign static threshold to the individual differences of heart failure patients becomes a difficult problem faced by the industry. Disclosure of Invention The application provides a method and a system for real-time monitoring vital signs of heart failure patients, which can realize accurate adaptation of a static threshold of vital signs to individual differences of heart failure patients. In a first aspect, the present application provides a method for real-time monitoring of vital signs of a heart failure patient, comprising the steps of: Acquiring real-time monitoring data of vital signs of a heart failure patient in a rehabilitation process; extracting corresponding physiological variability indexes from the real-time monitoring data of each vital sign, constructing an individual physiological baseline model according to all the physiological variability indexes and basic disease information of heart failure patients, and further generating a dynamic threshold interval of each vital sign; Performing cross-parameter association analysis on the deviation degree of the real-time monitoring data of each vital sign and the dynamic threshold interval based on a preset pathology association rule base to obtain pathology association indexes among different vital sign abnormal changes; Performing abnormality judgment on the real-time monitoring data of each vital sign according to all dynamic threshold intervals, and performing cross verification on an abnormality judgment result of each vital sign through the pathology correlation index to obtain an abnormality judgment confidence coefficient of each vital sign; and generating vital sign abnormality early warning information adapting to individual differences of heart failure patients when the abnormality judgment confidence degree of the vital signs meets a preset confidence condition. In some embodiments, extracting the corresponding physiological variability index from the real-time monitoring data of each vital sign specifically includes: Extracting time domain variation characteristics of real-time monitoring data of each vital sign; and determining the physiological variability index of each vital sign according to all the time domain variation characteristics. In some embodiments, constructing the personalized physiological baseline model from all physiological variability indices and underlying disease information of heart failure patients specifically includes: Acquiring basic disease information of heart failure patients, wherein the basic disease information comprises age, gender, duration of course, complications type and past hospitalization records; Constructing an initial physiological baseline model based on underlying diseas