Anesthesia recovery monitoring method and device and electronic equipment

Abstract

The invention provides an anesthesia recovery monitoring method, and the method comprises the steps: obtaining a first myoelectricity feature corresponding to a specified muscle group when a target object is in a non-anesthesia state, and obtaining a second myoelectricity feature corresponding to the specified muscle group when the target object is in a shallow anesthesia state, acquiring a third myoelectricity feature corresponding to a specified muscle group when the target object is in a deep anesthesia state; comprehensively comparing the third myoelectricity feature with the first myoelectricity feature and the second myoelectricity feature to obtain a feature comparison result of the specified muscle group; and determining whether the target object recovers from the anesthesia state or not according to the feature comparison result of the specified muscle group. The problems that an existing anesthesia recovery monitoring mode is not accurate enough in monitoring and unreliable in monitoring result are solved, and the accuracy of anesthesia recovery monitoring is improved.

Description

Technical Field

[0001] This invention relates to the field of medical analysis technology, and in particular to a method, device, and electronic device for monitoring anesthesia recovery. Background Technology

[0002] Post-anesthesia recovery monitoring is crucial for timely identification of patient safety. However, current methods often focus on collecting single physiological parameters to assess patient awakening and recovery risk, neglecting the adaptive optimization of sensor accuracy during monitoring. This can lead to unreliable results in subsequent assessments of patient awakening and recovery risk. Therefore, existing post-anesthesia recovery monitoring methods suffer from insufficient precision and unreliable results. Summary of the Invention

[0003] This invention provides a method for monitoring anesthesia recovery, aiming to address the problems of insufficient accuracy and unreliable results in existing anesthesia recovery monitoring methods. By acquiring the first electromyographic (EMG) characteristics of a specified muscle group in a non-anesthetized state, the second EMG characteristics of the same muscle group in a lightly anesthetized state, and the third EMG characteristics of the same muscle group in a deeply anesthetized state, the method comprehensively compares the third EMG characteristics with the first and second EMG characteristics to obtain the feature comparison results for the specified muscle group. Based on these results, it determines whether the target subject has recovered from anesthesia. This method solves the problems of insufficient accuracy and unreliable results in existing anesthesia recovery monitoring methods, thus improving the accuracy of anesthesia recovery monitoring.

[0004] In a first aspect, embodiments of the present invention provide a method for monitoring anesthesia recovery, the method comprising the following steps: The first electromyographic feature of a specified muscle group in a non-anesthetic state is obtained, the second electromyographic feature of the specified muscle group in a light anesthesia state is obtained, and the third electromyographic feature of the specified muscle group in a deep anesthesia state is obtained. The third electromyographic feature is compared with the first electromyographic feature and the second electromyographic feature to obtain the feature comparison result of the specified muscle group; Based on the feature comparison results of the specified muscle groups, it is determined whether the target object has recovered from the anesthesia state.

[0005] Optionally, obtaining the first electromyographic features corresponding to a specified muscle group of the target object in a non-anesthetized state includes: Acquire first electromyographic information corresponding to each specified muscle group when the target object performs a specified action in a non-anesthetized state, from a state of no movement to the completion of the specified action. Based on the first muscle synergy characteristics between the specified muscle groups under non-anesthesia conditions, a first feature extraction parameter corresponding to the muscle traction characteristics is matched to extract features from the first electromyographic information, thereby obtaining the first electromyographic features corresponding to the specified muscle groups under non-anesthesia conditions.

[0006] Optionally, obtaining the second electromyographic characteristics corresponding to the specified muscle group when the target object is under light anesthesia includes: Acquire second electromyographic information corresponding to each specified muscle group when the target object is in a state of light anesthesia, from no movement to the completion of a specified action; Based on the second muscle synergy characteristics between the specified muscle groups under light anesthesia, the second feature extraction parameters corresponding to the muscle traction characteristics are matched to extract features from the second electromyographic information, thereby obtaining the second electromyographic features corresponding to the specified muscle groups under light anesthesia.

[0007] Optionally, obtaining the third electromyographic features corresponding to the specified muscle group when the target object is under deep anesthesia includes: Real-time acquisition of third electromyographic information corresponding to each of the specified muscle groups of the target object under light anesthesia; Based on the third muscle synergy characteristics among the specified muscle groups under deep anesthesia, the third feature extraction parameters corresponding to the muscle traction characteristics are matched to extract the third electromyographic information in real time, thereby obtaining the third electromyographic features corresponding to the specified muscle groups under deep anesthesia.

[0008] Optionally, the step of comprehensively comparing the third electromyographic feature with the first and second electromyographic features to obtain the feature comparison result of the specified muscle group includes: The first electromyographic feature, the second electromyographic feature, and the third electromyographic feature are respectively encoded into a first electromyographic feature vector, a second electromyographic feature vector, and a third electromyographic feature vector representing a muscle synergy mode; Based on the first electromyographic feature vector and the second electromyographic feature vector, a dynamic "anesthesia-awake" state feature space is constructed, wherein the first electromyographic feature vector constitutes the cluster center of the awake state, the second electromyographic feature vector constitutes the cluster center of the light anesthesia state, and the decision boundary for state discrimination is determined based on the distribution relationship between the first electromyographic feature vector and the second electromyographic feature vector. The third electromyographic feature vector is mapped to the state feature space, and the feature comparison result of the specified muscle group is determined based on the distribution of the third electromyographic feature vector in the state feature space.

[0009] Optionally, determining the feature comparison result of the specified muscle group based on the distribution of the third electromyographic feature vector in the state feature space includes: Calculate the relative distance and convergence trend of the third electromyographic feature vector to the cluster centers of the conscious state and the light anesthesia state; Based on the motion trajectory of the third feature vector in the continuous time series, it is determined whether the decision boundary is crossed and converges to the awake state cluster center; Based on the analysis results of the relative distance, the approach trend, and the movement trajectory, the feature comparison results are generated, which include the degree to which the electromyographic pattern of the target object regresses from anesthesia to wakefulness.

[0010] Optionally, determining whether the target object has recovered from the anesthesia state based on the feature comparison results of the specified muscle group includes: If the degree to which the electromyographic pattern of the target object reverts from anesthesia to wakefulness is greater than or equal to a preset degree, then it is determined that the target object has recovered from the anesthesia. If the degree to which the electromyographic pattern of the target object returns from the anesthesia state to the awake state is less than a preset degree, then it is determined that the target object has not recovered from the anesthesia state.

[0011] Secondly, embodiments of the present invention also provide an anesthesia recovery monitoring device, the anesthesia recovery monitoring device comprising: The acquisition module is used to acquire the first electromyographic feature of a specified muscle group of the target object when the target object is in a non-anesthetic state, the second electromyographic feature of the specified muscle group of the target object when the target object is in a light anesthesia state, and the third electromyographic feature of the specified muscle group of the target object when the target object is in a deep anesthesia state. The feature comparison module is used to comprehensively compare the third electromyographic feature with the first electromyographic feature and the second electromyographic feature to obtain the feature comparison result of the specified muscle group. The judgment module is used to determine whether the target object has recovered from the anesthesia state based on the feature comparison results of the specified muscle group.

[0012] Thirdly, embodiments of the present invention provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the anesthesia recovery monitoring method provided in embodiments of the present invention.

[0013] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the anesthesia recovery monitoring method provided in embodiments of the present invention.

[0014] In this embodiment of the invention, the first electromyographic (EMG) feature corresponding to a specified muscle group in a non-anesthetic state, the second EMG feature corresponding to a specified muscle group in a lightly anesthetized state, and the third EMG feature corresponding to a specified muscle group in a deeply anesthetized state are obtained. The third EMG feature is then compared with the first and second EMG features to obtain the feature comparison result for the specified muscle group. Based on the feature comparison result of the specified muscle group, it is determined whether the target object has recovered from anesthesia. This invention solves the problems of insufficient accuracy and unreliable monitoring results in existing anesthesia recovery monitoring methods by obtaining the first EMG feature corresponding to a specified muscle group in a non-anesthetic state, the second EMG feature corresponding to a specified muscle group in a lightly anesthetized state, and the third EMG feature corresponding to a specified muscle group in a deeply anesthetized state, and by comparing the third EMG feature with the first and second EMG features to obtain the feature comparison result for the specified muscle group, and by determining whether the target object has recovered from anesthesia based on the feature comparison result of the specified muscle group. This improves the accuracy of anesthesia recovery monitoring. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, 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.

[0016] Figure 1 This is a flowchart of an anesthesia recovery monitoring method provided in an embodiment of the present invention; Figure 2 This is a raw electromyography (EMG) data graph; Figure 3 This is a graph showing the results of real-time characteristic value analysis between conscious and deep anesthesia. Figure 4 This is a real-time characteristic value analysis graph of deep anesthesia-awake state; Figure 5 This is a schematic diagram of the structure of an anesthesia recovery monitoring device provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] like Figure 1 As shown, Figure 1 This is a flowchart of an anesthesia recovery monitoring method provided by an embodiment of the present invention. The anesthesia recovery monitoring method includes the following steps: 101. Obtain the first electromyographic characteristics of a specified muscle group in a non-anesthetic state, the second electromyographic characteristics of a specified muscle group in a lightly anesthetized state, and the third electromyographic characteristics of a specified muscle group in a deep anesthetized state.

[0019] In this embodiment of the invention, the target object is a person who needs to be monitored for anesthesia recovery. The anesthesia recovery monitoring method is applied to a wearable device. The wearable device can be a carrier device for an electromyography (EMG) sensor, used to collect the wearer's EMG data. The EMG sensor can be an EMG sensor manufactured using fabric electrode technology. It adopts three-dimensional textile technology to realize a two-dimensional or three-dimensional mesh structure of different fiber blends, which can achieve better performance. Through human adaptation design, the textile fibers are digitized and used as electrodes for sampling EMG signals.

[0020] The aforementioned muscle groups can be the vastus medialis, rectus femoris, vastus lateralis, biceps femoris, semitendinosus, tibialis anterior, gastrocnemius, peroneus longus, and other muscle groups.

[0021] The aforementioned non-anesthetic state can be the physiological state of the target subject without receiving any anesthesia. The aforementioned light anesthesia state can be the anesthesia state in which the target subject can maintain consciousness and spontaneous breathing. The aforementioned deep anesthesia state can be the anesthesia state in which the target subject is unresponsive to painful stimuli, muscles are completely relaxed, and breathing and heart rate are slowed.

[0022] The first electromyographic feature mentioned above can be the electromyographic feature corresponding to a specified muscle group when the target subject is in a non-anesthetized state. The second electromyographic feature mentioned above can be the electromyographic feature corresponding to a specified muscle group when the target subject is in a lightly anesthetized state. The third electromyographic feature mentioned above can be the electromyographic feature corresponding to a specified muscle group when the target subject is in a deep anesthetized state.

[0023] The aforementioned electromyographic (EMG) characteristics can be obtained by sampling EMG signals from a specified muscle group using an EMG sensor under different anesthesia states. The EMG signals under different anesthesia states will exhibit different EMG characteristics. These characteristics include EMG amplitude, frequency, and duration.

[0024] It should be noted that the above-mentioned electromyographic signals are physiological electrical signals generated along with human muscle activity. They are electromyographic signals generated when the human body performs a specific action in pain or other situations.

[0025] 102. Compare the third electromyographic feature with the first and second electromyographic features to obtain the feature comparison results of the specified muscle group.

[0026] In this embodiment of the invention, the above-mentioned comprehensive comparison can be understood as a process of comprehensively comparing and analyzing the third electromyographic feature with the first and second electromyographic features to obtain the feature comparison results of a specified muscle group.

[0027] The above feature comparison results can be obtained by comparing and analyzing the third electromyographic feature with the first and second electromyographic features. These feature comparison results include the degree of regression of each specified muscle group from a deep anesthesia state to a conscious state.

[0028] It should be noted that the third, second, and first electromyographic features can be compared together to identify the characteristic changes of each electromyographic group under different anesthetic states.

[0029] 103. Based on the characteristic comparison results of the specified muscle groups, determine whether the target subject has recovered from anesthesia.

[0030] In this embodiment of the invention, it can be determined whether the target object has recovered from anesthesia based on the feature comparison results of a specified muscle group.

[0031] In some possible embodiments, when the feature comparison results of a specified muscle group show that the degree of regression of each specified muscle group from a deep anesthesia state to a conscious state is greater than or equal to the system's preset degree of regression from a deep anesthesia state to a conscious state, it can be determined that the target object has recovered from the anesthesia state.

[0032] In some possible embodiments, when the feature comparison results of the specified muscle groups show that the degree of regression of each specified muscle group from deep anesthesia to wakefulness is lower than the system's preset degree of regression from deep anesthesia to wakefulness, it can be determined that the target object has not recovered from the anesthesia state.

[0033] In this embodiment of the invention, the first electromyographic (EMG) feature corresponding to a specified muscle group in a non-anesthetic state, the second EMG feature corresponding to a specified muscle group in a lightly anesthetized state, and the third EMG feature corresponding to a specified muscle group in a deeply anesthetized state are obtained. The third EMG feature is then compared with the first and second EMG features to obtain the feature comparison result for the specified muscle group. Based on the feature comparison result of the specified muscle group, it is determined whether the target object has recovered from anesthesia. This invention solves the problems of insufficient accuracy and unreliable monitoring results in existing anesthesia recovery monitoring methods by obtaining the first EMG feature corresponding to a specified muscle group in a non-anesthetic state, the second EMG feature corresponding to a specified muscle group in a lightly anesthetized state, and the third EMG feature corresponding to a specified muscle group in a deeply anesthetized state, and by comparing the third EMG feature with the first and second EMG features to obtain the feature comparison result for the specified muscle group, and by determining whether the target object has recovered from anesthesia based on the feature comparison result of the specified muscle group. This improves the accuracy of anesthesia recovery monitoring.

[0034] It is understood that in the specific implementation of this application, data such as electromyography data, health data, and user data are involved. When the embodiments in this application are applied to specific products or technologies, user permission or consent is required. Furthermore, the collection, use, and processing of related data, as well as the training, deployment, and invocation of algorithm models, must comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0035] Optionally, in the step of obtaining the first electromyographic features corresponding to the specified muscle groups of the target object in a non-anesthetic state, the first electromyographic information corresponding to each specified muscle group can be obtained when the target object is in a non-anesthetic state from a state of no movement to the completion of the specified movement; based on the first muscle coordination characteristics between each specified muscle group in a non-anesthetic state, the first feature extraction parameters corresponding to the muscle traction features are matched to extract features from the first electromyographic information to obtain the first electromyographic features corresponding to the specified muscle groups in a non-anesthetic state.

[0036] In this embodiment of the invention, the aforementioned non-anesthetized state can be the physiological state of the target object without receiving any anesthesia.

[0037] The aforementioned inactive state can be a static state with no movement whatsoever. The aforementioned designated action can be a designated action such as clenching a fist. Specifically, the designated action can be performed by prompting the target object to execute it via voice, and the explanation of the designated action can be played aloud. Alternatively, staff can be prompted to give the target object the start instruction for the designated action, and the staff can explain how to complete the designated action.

[0038] The aforementioned first electromyographic information can be the electromyographic signals corresponding to each designated muscle group when the target object, in a non-anesthetic state, transitions from a state of inactivity to the completion of a specified action. This first electromyographic information can reflect the functional status of each muscle group in a non-anesthetic state.

[0039] The aforementioned muscle groups can be the vastus medialis, rectus femoris, vastus lateralis, biceps femoris, semitendinosus, tibialis anterior, gastrocnemius, peroneus longus, and other muscle groups.

[0040] The aforementioned first muscle synergy characteristic can be the synergistic effect formed when various muscle groups are simultaneously activated when performing a specified action from a state of inactivity.

[0041] The aforementioned muscle traction characteristics can be understood as the physical response of muscles during movement or under force, including extensibility and elasticity.

[0042] The first feature extraction parameter mentioned above can be a feature extraction parameter corresponding to the muscle traction feature. The first feature extraction parameter can be a set of values ​​or vectors that can represent the muscle traction feature in the absence of anesthesia.

[0043] The above feature extraction can be a process of extracting features from the first electromyographic information based on the first feature extraction parameters corresponding to the muscle traction features, and obtaining the first electromyographic features corresponding to the specified muscle group in a non-anesthetic state.

[0044] The aforementioned first electromyographic feature can be the electromyographic feature corresponding to a specific muscle group of the target object in a non-anesthetized tense state.

[0045] It should be noted that before the target object begins to perform the specified action, the target object keeps the target muscles relaxed and still, so that each muscle group is completely relaxed. Electromyography (EMG) sensors can be used to collect the first EMG information corresponding to each specified muscle group in the non-anesthetized state from the state of no movement to the completion of the specified action. Based on the first EMG synergistic characteristics between each muscle group, the first feature extraction parameters corresponding to the muscle traction characteristics are matched to extract features from the first EMG information, so as to obtain the first EMG features corresponding to the specified muscle group in the non-anesthetized state.

[0046] Optionally, in the step of obtaining the second electromyographic features corresponding to the specified muscle groups of the target object under light anesthesia, the second electromyographic information corresponding to each specified muscle group can be obtained when the target object is under light anesthesia, from a state of no movement to the completion of the specified movement; based on the second muscle coordination characteristics between each specified muscle group under light anesthesia, the second feature extraction parameters corresponding to the muscle traction features are matched to extract features from the second electromyographic information, thereby obtaining the second electromyographic features corresponding to the specified muscle groups under light anesthesia.

[0047] In this embodiment of the invention, the aforementioned light anesthesia state can be an anesthesia state in which the target subject can maintain consciousness and spontaneous breathing.

[0048] The aforementioned second electromyographic information can be the electromyographic signals corresponding to each designated muscle group when the target subject, under light anesthesia, transitions from a state of inactivity to the completion of a specified action. This second electromyographic information can reflect the functional status of each muscle group under light anesthesia.

[0049] The second muscle synergy characteristic can be the muscle synergy features between specific muscle groups under light anesthesia.

[0050] The aforementioned second feature extraction parameter can be a feature extraction parameter corresponding to the muscle traction feature. The second feature extraction parameter can represent a set of numerical values ​​or vectors of the muscle traction feature under light anesthesia. The above feature extraction can be a process of extracting features from the second electromyographic information based on the second feature extraction parameters corresponding to the muscle traction features, and obtaining the second electromyographic features corresponding to the specified muscle group under light anesthesia.

[0051] The aforementioned second electromyographic feature can be the electromyographic feature corresponding to a specific muscle group in the target subject under light anesthesia.

[0052] It should be noted that electromyography (EMG) sensors can be used to collect second EMG information corresponding to each designated muscle group during the transition from a state of no movement to the completion of a designated movement under light anesthesia. Based on the synergistic characteristics of the second EMG between each muscle group, second feature extraction parameters corresponding to the muscle traction characteristics are matched to extract features from the second EMG information, thereby obtaining the second EMG features corresponding to the designated muscle group under light anesthesia.

[0053] Optionally, in the step of obtaining the third electromyographic features corresponding to the specified muscle groups of the target object under deep anesthesia, the third electromyographic information corresponding to each specified muscle group of the target object under deep anesthesia can be obtained in real time; based on the third muscle synergy characteristics between each specified muscle group under deep anesthesia, the third feature extraction parameters corresponding to the muscle traction features are matched to extract the third electromyographic information in real time, so as to obtain the third electromyographic features corresponding to the specified muscle groups under deep anesthesia.

[0054] In this embodiment of the invention, the aforementioned deep anesthesia state can be an anesthesia state in which the target object does not respond to pain stimuli, the muscles are completely relaxed, and breathing and heartbeat are slowed down.

[0055] The aforementioned third electromyographic information can be the electromyographic information of the target subject under deep anesthesia, corresponding one-to-one with each designated muscle group. This third electromyographic information can reflect the functional status of each muscle group under deep anesthesia.

[0056] The aforementioned third muscle synergy characteristic can be the muscle synergy features between various designated muscle groups under deep anesthesia.

[0057] The third feature extraction parameter mentioned above can be the feature extraction parameter corresponding to muscle traction features under deep anesthesia.

[0058] The above feature extraction can be a process of extracting the third electromyographic information in real time based on the third feature extraction parameters corresponding to the muscle traction features, and obtaining the third electromyographic features corresponding to the specified muscle group under deep anesthesia.

[0059] The aforementioned third electromyographic feature can be the electromyographic feature corresponding to a specific muscle group in the target subject under deep anesthesia.

[0060] It should be noted that electromyography (EMG) sensors can be used to collect third EMG information corresponding to each designated muscle group during the transition from a state of no movement to the completion of a designated movement under deep anesthesia. Based on the third EMG synergistic characteristics between each muscle group, third feature extraction parameters corresponding to muscle traction characteristics are matched to extract features from the third EMG information, thereby obtaining the third EMG features corresponding to the designated muscle group under light anesthesia.

[0061] It should be noted that, considering the different electromyographic features implied in the electromyographic signals under different anesthesia states, different feature extraction parameters are set for different anesthesia states in this embodiment of the invention, thereby improving the accuracy of anesthesia recovery monitoring for different individuals.

[0062] Optionally, in the step of comprehensively comparing the third electromyographic feature with the first and second electromyographic features to obtain the feature comparison result of the specified muscle group, the first, second, and third electromyographic features can be encoded into first, second, and third electromyographic feature vectors respectively representing the muscle synergy mode; a dynamic "anesthesia-awake" state feature space is constructed based on the first and second electromyographic feature vectors; the third electromyographic feature vector is mapped to the state feature space, and the feature comparison result of the specified muscle group is determined based on the distribution of the third electromyographic feature vector in the state feature space.

[0063] In this embodiment of the invention, the first electromyographic feature vector constitutes a cluster center for the conscious state, and the second electromyographic feature vector constitutes a cluster center for the light anesthesia state. The decision boundary for state discrimination is determined based on the distribution relationship between the first and second electromyographic feature vectors. This decision boundary is used to distinguish between the conscious state and the light anesthesia state.

[0064] The aforementioned electromyographic feature vectors can reflect the activity of muscles.

[0065] The aforementioned muscle synergy pattern can be understood as a pattern in which various muscle groups participate in a specified movement simultaneously, achieving coordinated cooperation among the muscle groups.

[0066] The aforementioned "anesthesia-awake" state feature space is used to distinguish the anesthesia-awake state.

[0067] The above mapping can be a process of transforming the third electromyographic feature vector into a state feature space. Specifically, the third electromyographic feature vector can be projected into a dynamic "anesthesia-awake" state feature space to compare electromyographic features under different anesthesia states.

[0068] The above feature comparison results include changes in the specified muscle groups during the "anesthesia-awake" process.

[0069] It should be noted that dimensionality reduction techniques such as Principal Component Analysis (PCA) can be used to construct the state feature space. PCA, as mentioned above, maps high-dimensional data to a low-dimensional space through linear transformation while preserving the variance (information content) of the data. The goal of PCA is to find a new set of coordinate axes (principal components) that can capture the greatest variability in the data and approximate the original data with fewer dimensions.

[0070] Optionally, in the step of determining the feature comparison results of a specified muscle group based on the distribution of the third electromyographic feature vector in the state feature space, the relative distance and approach trend of the third electromyographic feature vector to the awake state cluster center and the light anesthesia state cluster center can be calculated; based on the motion trajectory of the third feature vector in the continuous time series, it can be determined whether it crosses the decision boundary and converges towards the awake state cluster center; based on the analysis results of the relative distance approach trend and motion trajectory, the feature comparison results can be generated.

[0071] In this embodiment of the invention, the relative distance from the third electromyographic feature vector to the cluster centers of the conscious state and the light anesthesia state can be calculated using Euclidean distance. The smaller the relative distance from the third electromyographic feature vector to the conscious state cluster center, the closer the third electromyographic feature vector is to the conscious state; conversely, the smaller the relative distance from the third electromyographic feature vector to the light anesthesia state cluster center, the closer the third electromyographic feature vector is to the light anesthesia state. The aforementioned Euclidean distance is the straight-line distance between two points in Euclidean space. A smaller Euclidean distance from the third electromyographic feature vector to the conscious state cluster center indicates a smaller relative distance between them, and vice versa.

[0072] The aforementioned convergence trend can be seen as the third electromyographic signal vector moving closer to the cluster centers of the conscious and clear states, and the cluster centers of the lightly anesthetized state. It can be understood that the smaller the relative distance between the third electromyographic feature vector and the cluster center of the conscious state, the closer it is to the conscious state; similarly, the smaller the relative distance between the third electromyographic feature vector and the cluster center of the lightly anesthetized state, the closer it is to the lightly anesthetized state, and so on.

[0073] The aforementioned decision boundary is used to distinguish between a conscious state and a state of light anesthesia.

[0074] The above feature comparison results include the degree to which the electromyographic patterns of the target object revert from anesthesia to wakefulness.

[0075] It should be noted that the relative distance and convergence trend of the third electromyographic feature vector to the cluster centers of the conscious state and the light anesthesia state can be calculated by Euclidean distance. Based on the trajectory of the third feature vector in the continuous time series, it can be determined whether the third feature vector crosses the decision boundary and converges to the cluster center of the conscious state. The feature comparison results are generated by analyzing the relative distance, convergence trend and trajectory.

[0076] In one possible implementation, for example, if the third electromyographic feature vector converges towards the cluster center of the conscious state, it is determined that the target user is moving from a state of deep anesthesia to a state of consciousness; if the third electromyographic feature vector converges towards the cluster center of the light anesthesia state, it is determined that the target user is moving from a state of deep anesthesia to a state of light anesthesia.

[0077] Optionally, in the step of determining whether the target object has recovered from anesthesia based on the characteristic comparison results of the specified muscle group, if the degree of regression of the target object's electromyographic pattern from anesthesia to wakefulness is greater than or equal to a preset degree, then the target object is determined to have recovered from anesthesia; if the degree of regression of the target object's electromyographic pattern from anesthesia to wakefulness is less than the preset degree, then the target object is determined not to have recovered from anesthesia.

[0078] In this embodiment of the invention, the aforementioned preset degree can be the degree to which the electromyographic pattern returns from anesthesia to wakefulness, which is preset by the system. It is used to determine whether the degree to which the electromyographic pattern of the target object returns from anesthesia to wakefulness reaches or exceeds the preset degree threshold. If it reaches or exceeds the preset degree, it can be determined that the target object has recovered from anesthesia. Conversely, if it does not reach the preset degree, it is determined that the target object has not recovered from anesthesia.

[0079] It should be noted that if the degree to which the target subject's electromyographic pattern returns from anesthesia to wakefulness is greater than or equal to a preset level, then the target subject is determined to have recovered from anesthesia; if the degree to which the target subject's electromyographic pattern returns from anesthesia to wakefulness is less than a preset level, then the target subject is determined not to have recovered from anesthesia.

[0080] Specifically, this invention uses an electromyography (EMG) sensor to sample EMG signals. The EMG signals are analog signals, which are converted into digital signals through analog-to-digital conversion. For example, the digital signals are values ​​between 0 and 1. In one possible embodiment, the first electrode samples the EMG signal of the vastus medialis muscle, with signals in channels 1 and 2; the second electrode samples the EMG signal of the rectus femoris muscle, with signals in channels 3 and 4; the third electrode samples the EMG signal of the vastus lateralis muscle, with signals in channels 5 and 6; the fourth electrode samples the EMG signal of the biceps femoris muscle, with signals in channels 7 and 8; the fifth electrode samples the EMG signal of the semitendinosus muscle, with signals in channels 9 and 10; the sixth electrode samples the EMG signal of the tibialis anterior muscle, with signals in channels 11 and 12; the seventh electrode samples the EMG signal of the gastrocnemius muscle, with signals in channels 13 and 14; and the eighth electrode samples the EMG signal of the peroneus longus muscle, with signals in channels 15 and 16.

[0081] like Figures 2-4 As shown, the example is taken to collect samples from the target subjects of anesthesia experiments. Figure 2This is a raw electromyography (EMG) data map. It was recorded from when the subject was awake until anesthesia, based on a fist-clenching command. The two EMG channels are located in the flexor muscles of the forearm (sensor 5) and the extensor muscles (sensor 7), respectively, more than three fingers' distance from the elbow. The experimental data shows that as anesthesia took effect, the amplitude of the EMG output from the subject's muscles gradually decreased. During the experiment, the subject struggled due to pain, resulting in a spike in amplitude. Figure 3 This is a graph showing the real-time characteristic value analysis results for awake-deep anesthesia. The amplitude of the channels on the flexor muscles significantly increases with fist clenching, with both RMS and IEMG increasing. As anesthesia takes effect, muscle strength decreases, reflected in a decrease in amplitude-related characteristic values ​​such as RMS. In the experiment, it was observed that the force output decreased progressively with the onset of anesthetic effects. The median frequency characteristic, with decreasing muscle activity, represents information related to muscle fatigue. After the patient became unable to respond to commands, MF rose to around 125Hz. It should be noted that by measuring MVC under awake conditions in the target subject and using this as a control, the anesthesia-awake status or the awake-anesthesia status can be assessed based on the RMS ratio. The percentage (RMS measurement ÷ RMS / VC × 100%) serves as a quantitative indicator for evaluating muscle recovery or the anesthesia process. Figure 4 This is a real-time characteristic value analysis chart of deep anesthesia-awakening. Specifically, it shows the data collected over a 9-minute real-time characteristic analysis period during the entire process from general anesthesia to recovery in the target subject. During the experiment, the amplitude values ​​shown in the chart remained largely unchanged, but the median frequency changed. The median frequency changes with the onset of anesthesia, and wavelet analysis can be used to analyze and compare the median frequency at each stage of anesthesia. Decreased MF: This is usually associated with muscle fatigue or the accumulation of metabolic products (such as lactic acid). During sustained muscle contraction, energy metabolism shifts from aerobic to anaerobic, leading to a decrease in local pH and muscle fiber conduction velocity, resulting in a reduction in high-frequency components and a leftward shift of MF. Increased MF: This indicates accelerated muscle recovery or metabolic clearance (such as post-anesthesia metabolic recovery), with high-frequency components regaining dominance. Anesthesia effects: By blocking neuromuscular junction transmission, it inhibits muscle contraction. During recovery, as metabolism and nerve impulse conduction resume, MF gradually recovers from low frequency to baseline levels.

[0082] like Figure 5 As shown, an embodiment of the present invention provides an anesthesia recovery monitoring device, which includes: The acquisition module 501 is used to acquire the first electromyographic feature of a specified muscle group when the target object is in a non-anesthetic state, the second electromyographic feature of the specified muscle group when the target object is in a light anesthesia state, and the third electromyographic feature of the specified muscle group when the target object is in a deep anesthesia state. The comprehensive comparison module 502 is used to comprehensively compare the third electromyographic feature with the first electromyographic feature and the second electromyographic feature to obtain the feature comparison result of the specified muscle group. The judgment module 503 is used to determine whether the target object has recovered from the anesthesia state based on the feature comparison results of the specified muscle group.

[0083] Optionally, the acquisition module 501 includes: The first acquisition submodule is used to acquire the first electromyographic information corresponding to each of the specified muscle groups when the target object is in a non-anesthetic state, from a state of no movement to the completion of a specified action. The first feature extraction submodule is used to extract features from the first electromyographic information by matching the first muscle synergy characteristics between the specified muscle groups under non-anesthesia conditions and the first feature extraction parameters corresponding to the muscle traction features, so as to obtain the first electromyographic features corresponding to the specified muscle groups under non-anesthesia conditions.

[0084] Optionally, the acquisition module 501 includes: The second acquisition submodule is used to acquire the second electromyographic information corresponding to each of the specified muscle groups when the target object is in a light anesthesia state, from a state of no movement to the completion of a specified movement. The second feature extraction submodule is used to extract features from the second electromyographic information by matching the second feature extraction parameters corresponding to the muscle traction features based on the second muscle synergy characteristics between the specified muscle groups under light anesthesia, thereby obtaining the second electromyographic features corresponding to the specified muscle groups under light anesthesia.

[0085] Optionally, the acquisition module 501 includes: The third acquisition submodule is used to acquire in real time the third electromyographic information of the target object corresponding to each of the specified muscle groups in a light anesthesia state. The third feature extraction submodule is used to extract features from the third electromyography information in real time by matching the third muscle synergy characteristics between the specified muscle groups under deep anesthesia and the third feature extraction parameters corresponding to the muscle traction features, so as to obtain the third electromyography features corresponding to the specified muscle groups under deep anesthesia.

[0086] Optionally, the comprehensive comparison module 502 includes: The encoding submodule is used to encode the first electromyographic feature, the second electromyographic feature, and the third electromyographic feature into a first electromyographic feature vector, a second electromyographic feature vector, and a third electromyographic feature vector representing a muscle synergy mode, respectively. A submodule is constructed to build a dynamic "anesthesia-awake" state feature space based on the first electromyography feature vector and the second electromyography feature vector. The first electromyography feature vector constitutes the cluster center of the awake state, and the second electromyography feature vector constitutes the cluster center of the light anesthesia state. The decision boundary for state discrimination is determined based on the distribution relationship of the first and second feature vectors. In the determination submodule, the user maps the third electromyographic feature vector to the state feature space, and determines the feature comparison result of the specified muscle group based on the distribution of the third electromyographic feature vector in the state feature space.

[0087] Optionally, the determining submodule includes: The calculation unit is used to calculate the relative distance and convergence trend of the third electromyographic feature vector to the cluster centers of the conscious state and the light anesthesia state; The judgment unit is used to determine whether the decision boundary is crossed based on the motion trajectory of the third feature vector in the continuous time series, and to converge towards the awake state cluster center; The generation unit is used to generate the feature comparison result based on the analysis results of the relative distance, the approaching situation and the motion trajectory. The feature comparison result includes the degree to which the electromyographic pattern of the target object regresses from anesthesia to wakefulness.

[0088] Optionally, the determination module 503 includes: The first judgment submodule is used to determine that the target object has recovered from the anesthesia if the degree to which the electromyographic pattern of the target object returns from the anesthesia state to the awake state is greater than or equal to a preset degree. The second judgment submodule is used to determine that the target object has not recovered from the anesthesia if the degree to which the electromyographic pattern of the target object returns from the anesthesia state to the awake state is less than a preset degree.

[0089] like Figure 6 As shown, this embodiment of the invention also provides an electronic device, including a processor, which can execute any of the above-described anesthesia recovery monitoring methods.

[0090] Specifically, it includes a processor 601 and a memory 602, as well as a computer program stored in the memory 602 and capable of running on the processor 601 to perform anesthesia recovery monitoring methods, wherein: The processor 601 runs the calculator program for the anesthesia recovery monitoring method stored in the memory 602, and performs the following steps: The first electromyographic feature of a specified muscle group in a non-anesthetic state is obtained, the second electromyographic feature of the specified muscle group in a light anesthesia state is obtained, and the third electromyographic feature of the specified muscle group in a deep anesthesia state is obtained. The third electromyographic feature is compared with the first and second electromyographic features to obtain the feature comparison result of the specified muscle group. Based on the feature comparison results of the specified muscle groups, it is determined whether the target object has recovered from the anesthesia state.

[0091] Optionally, the process of acquiring the first electromyographic features corresponding to a specified muscle group of the target object in a non-anesthetized state, executed by the processor 601, includes: Acquire first electromyographic information corresponding to each specified muscle group when the target object performs a specified action in a non-anesthetized state, from a state of no movement to the completion of the specified action. Based on the first muscle synergy characteristics between the specified muscle groups under non-anesthesia conditions, a first feature extraction parameter corresponding to the muscle traction characteristics is matched to extract features from the first electromyographic information, thereby obtaining the first electromyographic features corresponding to the specified muscle groups under non-anesthesia conditions.

[0092] Optionally, the step of processor 601 acquiring the second electromyographic features corresponding to the specified muscle group when the target object is under light anesthesia includes: Acquire second electromyographic information corresponding to each specified muscle group when the target object is in a state of light anesthesia, from no movement to the completion of a specified action; Based on the second muscle synergy characteristics between the specified muscle groups under light anesthesia, the second feature extraction parameters corresponding to the muscle traction characteristics are matched to extract features from the second electromyographic information, thereby obtaining the second electromyographic features corresponding to the specified muscle groups under light anesthesia.

[0093] Optionally, the process of obtaining the third electromyographic features corresponding to the specified muscle group when the target object is under deep anesthesia, executed by the processor 601, includes: Real-time acquisition of third electromyographic information corresponding to each of the specified muscle groups of the target object under light anesthesia; Based on the third muscle synergy characteristics among the specified muscle groups under deep anesthesia, the third feature extraction parameters corresponding to the muscle traction characteristics are matched to extract the third electromyographic information in real time, thereby obtaining the third electromyographic features corresponding to the specified muscle groups under deep anesthesia.

[0094] Optionally, the step of processor 601 performing a comprehensive comparison of the third electromyographic feature with the first and second electromyographic features to obtain the feature comparison result of the specified muscle group includes: The first electromyographic feature, the second electromyographic feature, and the third electromyographic feature are respectively encoded into a first electromyographic feature vector, a second electromyographic feature vector, and a third electromyographic feature vector representing the muscle synergy mode. Based on the first and second electromyographic feature vectors, a dynamic "anesthesia-awake" state feature space is constructed, wherein the first electromyographic feature vector constitutes the cluster center of the awake state, the second electromyographic feature vector constitutes the cluster center of the light anesthesia state, and the decision boundary for state discrimination is determined based on the distribution relationship of the first and second feature vectors. The third electromyographic feature vector is mapped to the state feature space, and the feature comparison result of the specified muscle group is determined based on the distribution of the third electromyographic feature vector in the state feature space.

[0095] Optionally, the processor 601's determination of the feature comparison result of the specified muscle group based on the distribution of the third electromyographic feature vector in the state feature space includes: Calculate the relative distance and convergence trend of the third electromyographic feature vector to the cluster centers of the conscious state and the light anesthesia state; Based on the motion trajectory of the third feature vector in the continuous time series, it is determined whether the decision boundary is crossed and converges to the awake state cluster center; Based on the analysis results of the relative distance, the approach trend, and the movement trajectory, the feature comparison results are generated, which include the degree to which the electromyographic pattern of the target object regresses from anesthesia to wakefulness.

[0096] Optionally, the step of determining whether the target object has recovered from the anesthesia state based on the feature comparison results of the specified muscle group, executed by the processor 601, includes: If the degree to which the electromyographic pattern of the target object reverts from anesthesia to wakefulness is greater than or equal to a preset degree, then it is determined that the target object has recovered from the anesthesia. If the degree to which the electromyographic pattern of the target object returns from the anesthesia state to the awake state is less than a preset degree, then it is determined that the target object has not recovered from the anesthesia state.

[0097] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the anesthesia recovery monitoring method provided in this invention and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0098] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0099] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for monitoring anesthesia recovery, characterized in that, The method includes the following steps: The first electromyographic feature of a specified muscle group in a non-anesthetic state is obtained, the second electromyographic feature of the specified muscle group in a light anesthesia state is obtained, and the third electromyographic feature of the specified muscle group in a deep anesthesia state is obtained. The third electromyographic feature is compared with the first electromyographic feature and the second electromyographic feature to obtain the feature comparison result of the specified muscle group; Based on the feature comparison results of the specified muscle groups, it is determined whether the target object has recovered from the anesthesia state.

2. The anesthesia recovery monitoring method as described in claim 1, characterized in that, The acquisition of the first electromyographic features corresponding to a specified muscle group of the target object in a non-anesthetized state includes: Acquire first electromyographic information corresponding to each specified muscle group when the target object performs a specified action in a non-anesthetized state, from a state of no movement to the completion of the specified action. Based on the first muscle synergy characteristics between the specified muscle groups under non-anesthesia conditions, a first feature extraction parameter corresponding to the muscle traction characteristics is matched to extract features from the first electromyographic information, thereby obtaining the first electromyographic features corresponding to the specified muscle groups under non-anesthesia conditions.

3. The anesthesia recovery monitoring method as described in claim 1, characterized in that, The step of obtaining the second electromyographic characteristics of the specified muscle group corresponding to the target object under light anesthesia includes: Acquire second electromyographic information corresponding to each specified muscle group when the target object is in a state of light anesthesia, from no movement to the completion of a specified action; Based on the second muscle synergy characteristics between the specified muscle groups under light anesthesia, the second feature extraction parameters corresponding to the muscle traction characteristics are matched to extract features from the second electromyographic information, thereby obtaining the second electromyographic features corresponding to the specified muscle groups under light anesthesia.

4. The anesthesia recovery monitoring method as described in claim 1, characterized in that, The step of obtaining the third electromyographic features corresponding to the specified muscle group of the target object under deep anesthesia includes: Real-time acquisition of third electromyographic information corresponding to each of the specified muscle groups of the target object under light anesthesia; Based on the third muscle synergy characteristics among the specified muscle groups under deep anesthesia, the third feature extraction parameters corresponding to the muscle traction characteristics are matched to extract the third electromyographic information in real time, thereby obtaining the third electromyographic features corresponding to the specified muscle groups under deep anesthesia.

5. The anesthesia recovery monitoring method according to any one of claims 1 to 4, characterized in that, The step of comprehensively comparing the third electromyographic feature with the first and second electromyographic features to obtain the feature comparison result of the specified muscle group includes: The first electromyographic feature, the second electromyographic feature, and the third electromyographic feature are respectively encoded into a first electromyographic feature vector, a second electromyographic feature vector, and a third electromyographic feature vector representing the muscle synergy mode. Based on the first electromyographic feature vector and the second electromyographic feature vector, a dynamic "anesthesia-awake" state feature space is constructed, wherein the first electromyographic feature vector constitutes the cluster center of the awake state, the second electromyographic feature vector constitutes the cluster center of the light anesthesia state, and the decision boundary for state discrimination is determined based on the distribution relationship between the first electromyographic feature vector and the second electromyographic feature vector. The third electromyographic feature vector is mapped to the state feature space, and the feature comparison result of the specified muscle group is determined based on the distribution of the third electromyographic feature vector in the state feature space.

6. The anesthesia recovery monitoring method as described in claim 5, characterized in that, The determination of the feature comparison results of the specified muscle group based on the distribution of the third electromyographic feature vector in the state feature space includes: Calculate the relative distance and convergence trend of the third electromyographic feature vector to the cluster centers of the conscious state and the light anesthesia state; Based on the motion trajectory of the third feature vector in the continuous time series, it is determined whether the decision boundary is crossed and converges to the awake state cluster center; Based on the analysis results of the relative distance, the approach trend, and the movement trajectory, the feature comparison results are generated, which include the degree to which the electromyographic pattern of the target object regresses from anesthesia to wakefulness.

7. The anesthesia recovery monitoring method as described in claim 6, characterized in that, The step of determining whether the target object has recovered from the anesthesia state based on the feature comparison results of the specified muscle group includes: If the degree to which the electromyographic pattern of the target object reverts from anesthesia to wakefulness is greater than or equal to a preset degree, then it is determined that the target object has recovered from the anesthesia. If the degree to which the electromyographic pattern of the target object returns from the anesthesia state to the awake state is less than a preset degree, then it is determined that the target object has not recovered from the anesthesia state.

8. An anesthesia recovery monitoring device, characterized in that, The anesthesia recovery monitoring device includes: The acquisition module is used to acquire the first electromyographic feature of a specified muscle group of the target object when the target object is in a non-anesthetic state, the second electromyographic feature of the specified muscle group of the target object when the target object is in a light anesthesia state, and the third electromyographic feature of the specified muscle group of the target object when the target object is in a deep anesthesia state. The feature comparison module is used to comprehensively compare the third electromyographic feature with the first electromyographic feature and the second electromyographic feature to obtain the feature comparison result of the specified muscle group. The judgment module is used to determine whether the target object has recovered from the anesthesia state based on the feature comparison results of the specified muscle group.

9. An electronic device, characterized in that, The electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the anesthesia recovery monitoring method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the anesthesia recovery monitoring method as described in any one of claims 1 to 7.

Images