CN-122004907-A - Debilitation risk assessment method and system based on lower limb myoelectricity and activation time sequence

CN122004907ACN 122004907 ACN122004907 ACN 122004907ACN-122004907-A

Abstract

The invention provides a weakening risk assessment method and a weakening risk assessment system based on lower limb myoelectricity and activation time sequences, and relates to the technical field of medical auxiliary detection and biological information processing. The method comprises the steps of obtaining historical lower limb surface myoelectricity data of a subject under a preset functional action, calculating and storing a normal interval of individual characteristic parameters, then collecting original surface myoelectricity signals of target muscle groups of the subject in a preset functional action period, carrying out digital filtering treatment, then carrying out multidimensional characteristic extraction, executing individual baseline correction, inputting multidimensional characteristics after the individual baseline correction into a preset cardiovascular debilitation risk assessment model, calculating a debilitation probability value, and outputting a risk assessment result. The invention realizes passive, quantitative and interpretable cardiovascular debilitation risk assessment in daily scenes, effectively corrects individual differences, and improves the accuracy of cardiovascular disease group debilitation identification and the intelligent level of clinical monitoring.

Inventors

CHEN YUAN
WU LINJING
HUANG PINGPING
LI JIAHUA
LIU SHUFEN
XUE MENGMENG
ZHANG RONGXIANG
ZHANG SHUJIE

Assignees

厦门大学附属心血管病医院(厦门市心脏中心)

Dates

Publication Date: 20260512
Application Date: 20260310

Claims (10)

1. A method for evaluating risk of weakness based on myoelectricity and activation timing of a lower limb, comprising: S1, acquiring historical lower limb surface myoelectricity data of a subject under a preset functional action, and calculating and storing a normal interval of individual characteristic parameters, wherein the preset functional action comprises at least three repeated sitting-standing conversion actions or walking actions; S2, acquiring original surface electromyographic signals of a target muscle group of a subject in a preset functional action period, and performing digital filtering treatment, wherein the target muscle group comprises right rectus femoris, right medial gastrocnemius and right tibialis anterior; S3, carrying out multidimensional feature extraction on the original surface electromyographic signals subjected to digital filtering treatment, wherein the multidimensional feature extraction comprises frequency domain feature extraction, activation time sequence feature extraction and clinical feature fusion; S4, combining the historical lower limb surface myoelectricity data, and executing individual baseline correction on the extracted multidimensional features; S5, inputting the multidimensional features corrected by the individual base lines into a pre-built cardiovascular debilitation risk assessment model, calculating a debilitation probability value, and outputting a risk assessment result according to a preset interval in which the debilitation probability value is located, wherein the cardiovascular debilitation risk assessment model is built by adopting a multivariate logistic regression equation.
2. The method for evaluating the risk of weakness based on the myoelectricity and activation time sequences of the lower limbs according to claim 1, wherein the digital filtering process obtains a signal envelope curve of the myoelectricity signal of the original surface by a full-wave rectification and low-pass filtering algorithm, and identifies the starting time of the muscle activation based on an envelope curve detection algorithm; The recognition criterion of the muscle activation starting time is that the signal amplitude exceeds twice standard deviation of the mean value of the rest section for the first time and the duration exceeds fifty milliseconds; the expression is as follows: ; ; ; Wherein, the -Providing said raw surface electromyographic signals; is a full wave rectified signal; is a low-pass filtered linear envelope; Is low-pass filtering; Is the mean value of the envelope of the rest state; is the rest state envelope standard deviation.
3. The method for evaluating the risk of weakness based on the myoelectricity and activation time sequences of the lower limbs according to claim 1, wherein the frequency domain feature extraction calculates the median frequency of the right rectus muscle and the right medial gastrocnemius muscle by performing fast fourier transform or power spectrum estimation on the segmented signals; The activation time sequence feature extraction is defined as an activation sequence index by calculating the time difference value of the right tibiofemoral muscle relative to the action starting event; the clinical features are fused by acquiring age data of a subject and a clinical function scale score, wherein the clinical function scale score adopts a balance and gait test scale.
4. The method of claim 1, wherein the individual baseline correction is to subtract the median frequency mean of individual characteristic parameters of the historical lower limb surface myoelectricity data from the extracted multidimensional characteristic and divide the median frequency mean by the standard deviation of individual characteristic parameters of the historical lower limb surface myoelectricity data to achieve characteristic normalization.
5. The method for evaluating the weakening risk based on the lower limb myoelectricity and the activation time sequence according to claim 1 is characterized in that when the original surface myoelectricity signals of specific lower limb myogroups of a subject are acquired in a preset functional action period, a scene triggering type data acquisition mode is adopted, namely, the acquisition is carried out through a pressure sensor array embedded below a seat cushion of a rehabilitation chair of the subject, when the pressure sensor array detects that the pressure value is reduced from a preset high threshold value corresponding to the weight of the subject to a preset low threshold value, the initiation of sitting-standing conversion action is judged, and the synchronous acquisition record of the original surface myoelectricity signals is started.
6. A method of evaluating risk of weakness based on myoelectricity and activation timing of lower limbs according to claim 3, wherein in the multivariate logistic regression equation of the cardiovascular risk evaluation model, the median frequency of right rectus and the median frequency of right medial gastrocnemius are used as forward predictors, the activation sequence index of the activation timing characteristic of right anterior tibial is used as reverse predictors, and fitting to the cardiovascular disease population specificity is achieved by increasing the weight of the frequency domain characteristic in the model.
7. The method for evaluating the risk of weakness based on the myoelectricity and activation time sequences of lower limbs according to claim 6, wherein the expression of the multivariate logistic regression equation is: ; Wherein, the Is the probability value of weakness; Is the intercept; 、、 Is a regression coefficient; Is the median frequency of the right rectus; is the median frequency of the right medial gastrocnemius; An activation sequence index that characterizes the activation timing of the right tibialis anterior.
8. The method for evaluating the risk of weakness based on the myoelectricity and activation sequences of lower limbs according to claim 1, further comprising the steps of generating a periodic report containing a characteristic change curve according to the risk evaluation result, and monitoring the risk of weakness of a subject in a cross-time dimension.
9. A weakening risk assessment system based on lower limb myoelectricity and activation timing, comprising: the fabric type lower limb myoelectricity acquisition protective clothing adopts an elastic fabric substrate, and a plurality of conductive fabric electrodes are integrated in the fabric type lower limb myoelectricity acquisition protective clothing and are used for being attached to a target muscle group part of a lower limb of a subject so as to acquire surface myoelectricity signals; The short-time trigger identification module comprises a pressure sensing unit embedded in a seat or an inertial measurement unit worn on a subject and is used for monitoring the action characteristics of the subject in real time and generating triggers; The portable data acquisition module is connected with the fabric type lower limb myoelectricity acquisition protector through a magnetic interface and is used for receiving trigger pulses and executing signal amplification, analog-to-digital conversion and local data caching; A processing unit and a remote server, wherein a computer program is stored in the remote server, and the computer program can be executed by the processing unit to realize a weakening risk assessment method based on lower limb myoelectricity and activation time sequence according to any one of claims 1-8; The user interaction interface is deployed at the mobile terminal or the medical workstation and is used for displaying the risk level and the history characteristic evolution trend of the subject.
10. The system for evaluating the risk of weakness based on the myoelectricity and activation sequences of the lower limbs according to claim 9, wherein the conductive fabric electrodes of the fabric-type lower limb myoelectricity acquisition protector are arranged in such a manner that a pair of differential main electrodes are respectively arranged at rectus femoris, semitendinosus/semimembranous zone, tibialis anterior and medial gastrocnemius positions, and a common reference electrode is arranged outside the ankle or knee joint; The conductive fabric electrode is manufactured through a silver fiber weaving process, and the surface of the conductive fabric electrode is covered with a conductive hydrogel layer.

Description

Debilitation risk assessment method and system based on lower limb myoelectricity and activation time sequence Technical Field The invention relates to the technical field of medical auxiliary detection and biological information processing, in particular to a weakening risk assessment method and system based on lower limb myoelectricity and activation time sequence. Background With the increasing global population aging, the identification of debilitation in elderly cardiovascular disease patients has become an important topic in the clinical monitoring and rehabilitation field. Debilitation is a typical senile syndrome manifested by a decline in physiological reserves and impaired multi-system function, and is clinically highly associated with falls, long hospitalization, loss of function, and an increased risk of death. Heart failure, coronary heart disease and other common cardiovascular diseases can obviously accelerate the occurrence process of weakness, so that early identification and risk assessment of the population are important. Currently, the identification of debilitation clinically depends mainly on scale questionnaires or basic functional tests, such as Fried debilitation phenotype scoring, ten meter pace testing, balance function assessment, and the like. Although the methods are widely applied in clinic, the methods have the limitations of strong subjectivity, dependence on the coordination degree of a subject, lack of objective quantitative physiological markers and the like, and particularly are difficult to realize passive and convenient monitoring in the daily functional actions of the elderly, such as the switching of sitting and standing or starting process. The surface electromyography (sEMG) technique is used as a non-invasive muscle electrical activity recording means, and can effectively reflect the muscle activation intensity, the muscle fatigue and the neuromuscular control characteristics. However, the existing surface myoelectricity research focuses on short-time testing in laboratory environment, and feature extraction focuses on single time domain indexes such as root mean square value, integral myoelectricity and the like, neglecting frequency domain features such as median frequency, average power frequency and activation time sequence among muscle groups, and evaluating deep discrimination value in neuromuscular coordination capacity and cardiovascular disease related weakness. The existing biological signal monitoring patent scheme relates to the fusion of multi-mode data, but does not conduct right limitation and model construction aiming at the independent prediction value of the frequency domain and time sequence characteristics of specific muscle groups of lower limbs in cardiovascular disease groups. In addition, the prior art presents a number of challenges in clinical applications. Firstly, the limitation of the acquisition scene is that the system is difficult to be integrated into the environment of a pension institution or a ward without sense due to the lack of a trigger-type acquisition scheme bound with daily functional actions. Secondly, the model lacks pertinence, and the muscular electrophysiology performance of patients with cardiovascular diseases is obviously different from that of general elderly people due to special cardiogenic metabolic changes and abnormal muscular perfusion, and the general model often causes assessment misalignment. Furthermore, the existing myoelectricity acquisition system is mostly dependent on a disposable patch electrode and a complex manual labeling process, has poor clinical portability and complex operation, and is difficult to popularize in home follow-up or bedside routine monitoring. Finally, existing assessment methods are not robust enough in the face of significant individual differences due to the lack of efficient individual baseline correction and threshold self-adaptation mechanisms. In view of this, the present application has been made. Disclosure of Invention The invention aims to provide a weakening risk assessment method and a weakening risk assessment system based on lower limb myoelectricity and activation time sequences, which are used for solving the defects that in the existing method, weakening identification of old cardiovascular disease patients excessively depends on subjective questionnaires, convenient and objective quantitative physiological markers in daily scenes are lacked, and the existing surface myoelectricity monitoring scheme is lacked a special model for cardiovascular disease groups and is insufficient in individual difference correction. In order to solve the technical problems, the invention is realized by the following technical scheme: a method for evaluating risk of weakness based on lower limb myoelectricity and activation timing, comprising: S1, acquiring historical lower limb surface myoelectricity data of a subject under a preset functional action, and calculating and storin