Search

CN-122004906-A - Myoelectric signal-based myorelaxation depth prediction anesthesia system

CN122004906ACN 122004906 ACN122004906 ACN 122004906ACN-122004906-A

Abstract

The embodiment of the invention provides a myoelectric signal-based myorelaxant depth prediction anesthesia system, and relates to the technical field of medical monitoring technology. The system comprises an electromyographic signal acquisition module, a signal preprocessing module, a spectrum state analysis module and a trend prediction and display module, wherein the electromyographic signal acquisition module is used for acquiring electromyographic signals, the signal preprocessing module is used for preprocessing the electromyographic signals, the spectrum state analysis module is used for carrying out windowing and spectrum transformation on the preprocessed electromyographic signals so as to generate spectrum characteristics of each time window, and calculating spectrum inertia reflecting dynamic changes of the spectrum characteristics based on a preset spectrum inertia model, wherein the spectrum characteristics comprise median frequency and total power, and the trend prediction and display module is used for combining the current muscle relaxation depth and the spectrum inertia to generate and display a muscle relaxation trend prediction result. The invention solves the problem of low reliability of the muscle relaxation depth, thereby achieving the effect of improving the prediction precision.

Inventors

  • JIANG HUIFANG

Assignees

  • 浙江省肿瘤医院

Dates

Publication Date
20260512
Application Date
20260205

Claims (10)

  1. 1. A myoelectric signal-based myorelaxant depth prediction anesthesia system, comprising: The myoelectric signal acquisition module is used for acquiring myoelectric signals; the signal preprocessing module is used for preprocessing the electromyographic signals; The spectrum state analysis module is used for carrying out window division and spectrum transformation on the preprocessed electromyographic signals to generate spectrum characteristics of each time window, and calculating spectrum inertia reflecting dynamic changes of the spectrum characteristics based on a preset spectrum inertia model, wherein the spectrum characteristics comprise median frequency and total power; and the trend prediction and display module is used for combining the current muscle relaxation depth and the frequency spectrum inertia to generate and display a muscle relaxation trend prediction result.
  2. 2. The system of claim 1, further comprising: before the step of carrying out windowing and spectrum transformation on the preprocessed electromyographic signals to generate spectrum characteristics of each time window, acquiring baseline electromyographic signals when the onset of myorelaxant reaches a peak value; And determining the baseline median frequency and the baseline total power based on the baseline electromyographic signals.
  3. 3. The system of claim 2, further comprising: for a current time window, constructing a normalized spectral state vector based on the median frequency, the total power, the baseline median frequency, and the baseline total power; calculating a spectrum speed vector based on the normalized spectrum state vector of the current time window and the normalized spectrum state vector of the last time window; the spectral inertia is updated by exponentially weighted moving averages of norms of the spectral velocity vectors.
  4. 4. The system of claim 1, wherein the generating a muscle relaxation trend prediction result by combining the current muscle relaxation depth and the spectral inertia comprises: comparing the frequency spectrum inertia with a preset inertia threshold value; If the frequency spectrum inertia is greater than or equal to the inertia threshold, judging that the muscle relaxation state is in a change and generating a corresponding muscle relaxation trend prediction result, or distinguishing the type of muscle relaxation recovery based on the component direction of the frequency spectrum speed vector, wherein if each component of the frequency spectrum speed vector is positive, judging that the muscle relaxation trend prediction result is fast recovery; And if the frequency spectrum inertia is smaller than the inertia threshold, judging that the muscle relaxation state is stable, and generating a corresponding muscle relaxation trend prediction result.
  5. 5. The myoelectric signal-based myorelaxant depth prediction anesthesia method is characterized by comprising the following steps of: collecting and preprocessing electromyographic signals; Windowing and spectrum transformation are carried out on the preprocessed electromyographic signals so as to generate spectrum characteristics of each time window, wherein the spectrum characteristics comprise median frequency and total power; Calculating the frequency spectrum inertia reflecting the dynamic change of the frequency spectrum characteristics based on a preset frequency spectrum inertia model; And generating a predicted result of the muscle relaxation trend by combining the current muscle relaxation depth and the frequency spectrum inertia.
  6. 6. The method of claim 5, wherein prior to said windowing and spectral transforming the pre-processed electromyographic signals to generate spectral features for each time window, the method further comprises: When the onset of the muscle relaxant reaches a peak value, acquiring a baseline myoelectric signal; And determining the baseline median frequency and the baseline total power based on the baseline electromyographic signals.
  7. 7. The method of claim 6, wherein the step of calculating the spectral inertia reflecting the dynamic change of the spectral features based on a predetermined spectral inertia model comprises: for a current time window, constructing a normalized spectral state vector based on the median frequency, the total power, the baseline median frequency, and the baseline total power; calculating a spectrum speed vector based on the normalized spectrum state vector of the current time window and the normalized spectrum state vector of the last time window; Updating the spectral inertia based on the spectral velocity vector.
  8. 8. The method of claim 5, wherein the generating a muscle relaxation trend prediction result by combining the current muscle relaxation depth and the spectral inertia comprises: comparing the frequency spectrum inertia with a preset inertia threshold value; If the frequency spectrum inertia is greater than or equal to the inertia threshold, judging that the muscle relaxation state is in a change and generating a corresponding muscle relaxation trend prediction result, or distinguishing the type of muscle relaxation recovery based on the component direction of the frequency spectrum speed vector, wherein if each component of the frequency spectrum speed vector is positive, judging that the muscle relaxation trend prediction result is fast recovery; And if the frequency spectrum inertia is smaller than the inertia threshold, judging that the muscle relaxation state is stable, and generating a corresponding muscle relaxation trend prediction result.
  9. 9. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program, wherein the computer program is arranged to execute the method of any of the claims 5 to 8 when run.
  10. 10. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the method of any of the claims 5 to 8.

Description

Myoelectric signal-based myorelaxation depth prediction anesthesia system Technical Field The embodiment of the invention relates to the field of medical monitoring, in particular to a myoelectric signal-based myorelaxation depth prediction anesthesia system. Background In general anesthesia operation, it is important to accurately regulate the muscle relaxation level of a patient, and excessive muscle relaxation may cause body movement of the patient during operation, interfere with operation progress, even cause severe complications such as choking cough and laryngeal spasm, while excessive muscle relaxation may prolong postoperative recovery time and increase risks such as respiratory depression. Therefore, the method is a core link for guaranteeing anesthesia safety by continuously and accurately monitoring the muscle relaxation depth and pre-judging the future change trend. Currently, the clinical dependence on stimulation of peripheral nerves and observation of the contractile response of the corresponding muscles is mainly used to evaluate muscle relaxation, such as four-Train-of-Four (TOF) counts. However, such methods are typically intermittent, do not provide continuous monitoring data, and in the deep muscle relaxation state (TOF count of 0), the resolution and sensitivity are significantly reduced. To achieve continuous monitoring, academia and industry began to study the use of patient spontaneous myoelectric signals to assess muscle relaxation. Some approaches attempt to analyze the time series characteristics of the electromyographic signals using a machine learning model, such as the long short term memory network (LSTM), in an effort to predict the trend of muscle relaxation. However, the 'black box' characteristic of the model makes the decision process difficult to explain, and the clinician cannot intuitively understand the basis of the prediction result, so that the credibility of the model in the key medical decision is limited. Disclosure of Invention The embodiment of the invention provides a myoelectric signal-based myorelaxant depth prediction anesthesia system, which at least solves the problem of low decision reliability in the related technology. According to one embodiment of the present invention, there is provided a myoelectric signal-based myorelaxant depth prediction anesthesia system including: The myoelectric signal acquisition module is used for acquiring myoelectric signals; the signal preprocessing module is used for preprocessing the electromyographic signals; The spectrum state analysis module is used for carrying out window division and spectrum transformation on the preprocessed electromyographic signals to generate spectrum characteristics of each time window, and calculating spectrum inertia reflecting dynamic changes of the spectrum characteristics based on a preset spectrum inertia model, wherein the spectrum characteristics comprise median frequency and total power; and the trend prediction and display module is used for combining the current muscle relaxation depth and the frequency spectrum inertia to generate and display a muscle relaxation trend prediction result. In one exemplary embodiment, further comprising: before the step of carrying out windowing and spectrum transformation on the preprocessed electromyographic signals to generate spectrum characteristics of each time window, acquiring baseline electromyographic signals when the onset of myorelaxant reaches a peak value; And determining the baseline median frequency and the baseline total power based on the baseline electromyographic signals. In one exemplary embodiment, further comprising: for a current time window, constructing a normalized spectral state vector based on the median frequency, the total power, the baseline median frequency, and the baseline total power; calculating a spectrum speed vector based on the normalized spectrum state vector of the current time window and the normalized spectrum state vector of the last time window; the spectral inertia is updated by exponentially weighted moving averages of norms of the spectral velocity vectors. In an exemplary embodiment, the generating the predicted muscular relaxation tendency result includes: comparing the frequency spectrum inertia with a preset inertia threshold value; If the frequency spectrum inertia is greater than or equal to the inertia threshold, judging that the muscle relaxation state is in a change and generating a corresponding muscle relaxation trend prediction result, or distinguishing the type of muscle relaxation recovery based on the component direction of the frequency spectrum speed vector, wherein if each component of the frequency spectrum speed vector is positive, judging that the muscle relaxation trend prediction result is fast recovery; And if the frequency spectrum inertia is smaller than the inertia threshold, judging that the muscle relaxation state is stable, and generating a corresponding muscle relaxati