Electrical stimulation device, method of controlling an electrical stimulation device

Abstract

The present disclosure provides an electrical stimulation device and a control method thereof. The electrical stimulation device comprises: an electrode, which is in contact with a target object in a use state, for electrically stimulating a target tissue of the target object; a sensor, which is used to collect respiratory data of the target object, the respiratory data comprising at least one of: motion data, exhalation temperature data, thoracic pressure data, electrical impedance data, blood oxygen data, and posture data; and a controller, which is electrically connected with the sensor and the electrode, and is used to: determine a predicted sleep state of the target object based on the respiratory data collected by the sensor; determine an output parameter of a stimulation signal based on the predicted sleep state; and control the electrode to output the stimulation signal based on the output parameter of the stimulation signal, so as to stimulate the target tissue. The sleep quality of the target object when wearing the electrical stimulation device is improved, and the stimulation effect of the stimulation signal is improved.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of nerve stimulation, and in particular to an electrical stimulation device and a control method of the electrical stimulation device. BACKGROUND

[0002] A user with sleep breathing disorder will cause the pharyngeal and lingual muscles to relax physiologically during sleep, resulting in the tongue blocking the pharyngeal cavity passage at the back during sleep, and further causing respiratory pause and low ventilation during sleep. Therefore, it is necessary to stimulate the nerves or muscles of the patient with sleep breathing disorder to enhance and maintain the openness of the upper respiratory tract of the patient, so as to reduce or eliminate the airway obstruction caused by the backward falling of the root of the tongue.

[0003] In the related art, the patient will turn on the electrical stimulation device before falling asleep. The electrical stimulation device will directly start outputting the stimulation signal in a specific mode after being turned on, and will stop outputting the stimulation signal only when the patient manually turns off the electrical stimulation device after ending sleep. However, the stimulation signal output by the electrical stimulation device will affect the sleep state of the patient, and the fixed stimulation mode is difficult to automatically adapt to the change of the sleep time of the patient, and does not have uniform adaptability among different patients, thus seriously affecting the sleep quality of the user. SUMMARY

[0004] Therefore, the present disclosure provides an electrical stimulation device and a control method of the electrical stimulation device, which can effectively improve the output time of the electrical stimulation signal and improve the user experience.

[0005] According to a first aspect of an embodiment of the present disclosure, an electrical stimulation device is provided, which comprises:

[0006] an electrode, which is in contact with a target object in a use state, and is used for electrical stimulation of a target tissue of the body of the target object;

[0007] a sensor, which is used for collecting breathing data of the target object, the breathing data comprising at least one of the following: motion data, exhalation temperature data, thoracic pressure data, electrical impedance data, blood oxygen data, and posture data;

[0008] a controller, which is electrically connected with the sensor and the electrode, and is used for:

[0009] determining a predicted sleep state of the target object based on the breathing data collected by the sensor;

[0010] determining an output parameter of a stimulation signal based on the predicted sleep state;

[0011] controlling the electrode to output the stimulation signal based on the output parameter of the stimulation signal, so as to stimulate the target tissue.

[0012] In some embodiments, determining the predicted sleep state of the target object based on the respiration data collected by the sensor comprises:

[0013] determining a predicted respiration state of the target object in a next respiration cycle according to the respiration data;

[0014] determining a sleep posture signal of the target object according to posture data and electrical impedance data in the respiration data;

[0015] determining a sleep movement signal of the target object according to movement data and thoracic pressure data in the respiration data;

[0016] determining the predicted sleep state of the target object according to blood oxygen data, the predicted respiration state, the sleep posture signal and the sleep movement signal in the respiration data.

[0017] In some embodiments, determining the predicted respiration state of the target object in a next respiration cycle based on the respiration data comprises:

[0018] determining a current respiration state of the target object according to the respiration data;

[0019] determining the predicted respiration state of the target object in a next respiration cycle according to the current respiration state of the target object.

[0020] In some embodiments, determining the current respiration state of the target object according to the respiration data comprises:

[0021] in response to determining that there is an interference signal in a respiration signal of the target object, determining a sampling window according to a preset respiration frequency range and a preset respiration movement feature, wherein the respiration signal is determined according to movement data in the respiration data;

[0022] performing feature extraction on the respiration signal in the sampling window according to the preset respiration movement feature to determine a feature weight;

[0023] determining a simulated respiration signal according to a reference respiration frequency and a reference respiration phase;

[0024] adjusting the reference respiration frequency and the reference respiration phase according to the feature weight to minimize a residual error between the respiration signal and the simulated respiration signal, and determining a current respiration frequency and a current respiration phase in the current respiration state.

[0025] In some embodiments, determining the predicted respiration state of the target object in a next respiration cycle according to the current respiration state of the target object comprises:

[0026] based on a pre-constructed respiratory state space model;

[0027] perform feature extraction on the current respiratory state to obtain a feature vector;

[0028] based on the state vector corresponding to the current respiratory state and the feature vector, predict a predicted respiratory state of the target object in a next respiratory cycle through the state space model.

[0029] In some embodiments, the determining, according to the predicted sleep state, an output parameter of a stimulation signal comprises:

[0030] in response to the predicted sleep state representing that the target object starts to enter sleep, determining the output parameter of the stimulation signal according to sleep data of the target object in a historical time window;

[0031] in response to the predicted sleep state representing that the target object is in a wake state, stopping outputting the stimulation signal.

[0032] In some embodiments, the controller is further configured to:

[0033] determine a correction function of the sleep respiratory model based on a difference between an actual sleep state of the target object and the predicted sleep state;

[0034] correct the sleep respiratory model based on the correction function.

[0035] In some embodiments, the controller is further configured to:

[0036] update the respiratory state space model based on a difference between an actual respiratory state of the target object in a next respiratory cycle and the predicted respiratory state;

[0037] adjust a collection parameter of the sensor based on the predicted respiratory state, wherein the collection parameter of the sensor comprises at least one of a collection frequency, a resolution, and a power.

[0038] According to a second aspect of the embodiments of the present disclosure, a control method of an electrical stimulation device is provided, the method comprising:

[0039] acquiring, by a sensor, respiratory data of a target object, wherein the respiratory data comprises at least one of motion data, thoracic pressure data, electrical impedance data, blood oxygen data, and posture data;

[0040] determining, by a controller, a predicted sleep state of the target object based on the respiratory data;

[0041] determining, by the controller, an output parameter of the stimulation signal based on the predicted sleep state;

[0042] controlling, by the controller, the electrode to output the stimulation signal based on the output parameter of the stimulation signal.

[0043] In some embodiments, the determining, by the processor, the predicted sleep state of the target subject based on the respiration data comprises:

[0044] determining, by the processor, a predicted respiration state of the target subject in a next respiration cycle based on the respiration data;

[0045] determining, by the processor, a sleep posture signal of the target subject based on posture data and electrical impedance data in the respiration data;

[0046] determining, by the processor, a sleep movement signal of the target subject based on movement data and thoracic pressure data in the respiration data;

[0047] determining, by the processor, the predicted sleep state of the target subject based on blood oxygen data, the predicted respiration state, the sleep posture signal and the sleep movement signal in the respiration data.

[0048] the determining, by the processor, the predicted respiration state of the target subject in a next respiration cycle based on the respiration data comprises:

[0049] determining, by the processor, a current respiration state of the target subject based on the respiration data;

[0050] determining, by the processor, the predicted respiration state of the target subject in a next respiration cycle based on the current respiration state of the target subject.

[0051] In some embodiments, the determining, by the processor, the current respiration state of the target subject based on the respiration data comprises:

[0052] in response to determining that there is an interference signal in the respiration signal of the target subject, determining a sampling window based on a preset respiration frequency range and a preset respiration movement feature, wherein the respiration signal is determined based on movement data in the respiration data;

[0053] performing feature extraction on the respiration signal in the sampling window based on the preset respiration movement feature to determine a feature weight;

[0054] determining a simulated respiration signal based on a reference respiration frequency and a reference respiration phase;

[0055] adjusting the reference respiration frequency and the reference respiration phase based on the feature weight to minimize a residual error between the respiration signal and the simulated respiration signal, and determining a current respiration frequency and a current respiration phase in the current respiration state.

[0056] In some embodiments, the determining the predicted respiratory state of the target object in the next respiratory cycle according to the current respiratory state of the target object comprises:

[0057] based on a pre-constructed respiratory state space model;

[0058] performing feature extraction on the current respiratory state to obtain a feature vector;

[0059] predicting the predicted respiratory state of the target object in the next respiratory cycle based on the state vector corresponding to the current respiratory state and the feature vector through the state space model.

[0060] In some embodiments, the determining the output parameter of the stimulation signal according to the predicted sleep state comprises:

[0061] in response to the predicted sleep state representing that the target object starts to enter sleep, determining the output parameter of the stimulation signal according to sleep data of the target object in a historical time window;

[0062] in response to the predicted sleep state representing that the target object is in a wake state, stopping outputting the stimulation signal.

[0063] In some embodiments, the method further comprises:

[0064] determining a correction function of the respiratory state space model based on a difference between the real-time respiratory state of the target object in the next respiratory cycle and the predicted respiratory state;

[0065] correcting the respiratory state space model based on the correction function.

[0066] In some embodiments, the controller is further configured to:

[0067] updating the respiratory state space model based on a difference between the actual respiratory state of the target object in the next respiratory cycle and the predicted respiratory state;

[0068] adjusting the acquisition parameter of the sensor based on the predicted respiratory state, wherein the acquisition parameter of the sensor comprises at least one of the following: acquisition frequency, resolution and power.

[0069] The technical solutions provided by the embodiments of the present disclosure can have the following beneficial effects:

[0070] The respiration data of the target object is collected through the sensor, so that the controller determines the predicted sleep state of the target object based on the respiration data, and then determines the output parameter of the stimulation signal according to the predicted sleep state, to control the electrode to output the stimulation signal. The purpose of self-adaptive adjustment of the stimulation signal of the electric stimulation device according to the sleep state of the target object can be achieved, the target object is prevented from being disturbed by the stimulation signal in the sleep stage, the sleep quality of the target object when wearing the electric stimulation device is improved, and then the stimulation effect of the stimulation signal is improved.

[0071] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0072] The accompanying drawings, which are incorporated into the specification and constitute part of it, illustrate embodiments consistent with the present disclosure and, together with the specification, serve to explain the principles of the present disclosure.

[0073] Figure 1 is a structure diagram of an electric stimulation device according to an exemplary embodiment of the present disclosure;

[0074] Figure 2 is a flowchart of determining a predicted sleep state according to an exemplary embodiment of the present disclosure;

[0075] Figure 3 is a flowchart of determining a respiration frequency and phase according to an exemplary embodiment of the present disclosure;

[0076] Figure 4 is a flowchart of determining a predicted respiration state according to an exemplary embodiment of the present disclosure;

[0077] Figure 5 is a flowchart of a determination method of an electric stimulation signal according to an exemplary embodiment of the present disclosure;

[0078] Figure 6 is a flowchart of a determination method of an electric stimulation signal according to an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0079] The exemplary embodiments will be described in detail hereinafter with reference to the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present disclosure. Instead, they are merely examples of apparatuses and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0080] The terminology used in the present disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the present disclosure. As used in the present disclosure and the appended claims, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0081] It should be understood that although the terms first, second, third, etc. can be employed in this disclosure to describe various information, these information should not be limited to these terms. These terms are only used to differentiate one piece of information from another piece of information of the same type. For example, a first information can also be referred to as a second information, and similarly, a second information can also be referred to as a first information, without departing from the scope of the present disclosure. Depending on the context, the word "if' as used herein can be interpreted to mean "when" or "upon" or "in response to determining".

[0082] For a user with sleep breathing disorder, the muscle tissue of the tongue or the hypoglossal nerve of the user can be electrically stimulated by an electrical stimulation device during the sleep of the user to make the target tissue of the user contract and keep the upper respiratory tract smooth. In the related art, the user usually wears and turns on the electrical stimulation device when preparing to sleep, so that the electrical stimulation device can output a stimulation signal. However, after the user prepares to sleep, the user does not enter the sleep state immediately, and if the stimulation signal is output at the accurate sleep stage of the user, the stimulation signal will cause a flu-like condition in the stimulated area or even cause discomfort, affect the normal sleep process of the user, and increase the sleep time. Moreover, when the user wakes up again or gets up after entering the sleep state, the stimulation signal will also affect the user to enter the sleep state again, resulting in a rejection phenomenon to the stimulation treatment and reducing the sleep quality of the user.

[0083] Based on this, the electrical stimulation device provided in the embodiments of the present disclosure can be a placement type electrical stimulation device, specifically, a tongue muscle stimulator, which electrically stimulates the muscle tissue of the tongue of the user by being placed in the oral cavity of the user. In addition, it can also be an implantable electrical stimulation device, specifically, a hypoglossal nerve stimulator, which electrically stimulates the hypoglossal nerve of the user by being implanted in the human body of the user. The electrical stimulation device can adjust the output of the stimulation signal to the muscle tissue of the tongue or the hypoglossal nerve of the user according to the sleep state of the user, so as to make the user keep the respiratory tract smooth in the sleep state while ensuring the sleep quality of the user. The electrical stimulation device comprises:

[0084] an electrode, which is in contact with a target object in a use state, and is used to electrically stimulate a target tissue of the target object;

[0085] a sensor configured to collect breathing data of the target object, the breathing data comprising at least one of: motion data, exhalation temperature data, thoracic pressure data, electrical impedance data, blood oxygen data, and posture data;

[0086] a controller in electrical connection with the sensor and the electrode, configured to:

[0087] determine a predicted sleep state of the target object based on the breathing data collected by the sensor;

[0088] determine an output parameter of the stimulation signal based on the predicted sleep state;

[0089] control the electrode to output the stimulation signal based on the output parameter of the stimulation signal to stimulate the target tissue.

[0090] In some embodiments, the structure diagram in the electrical stimulation device can be as shown in Figure 1 The controller is in electrical connection with the electrode and the sensor, respectively. The sensor collects breathing data of the target object during sleep and transmits the breathing data to the controller. The controller determines the predicted sleep state of the target object based on the breathing data collected by the sensor, and then determines the output parameter of the stimulation signal for stimulating the tongue muscle or the hypoglossal nerve of the target object based on the predicted sleep state of the target object. Finally, the electrode outputs the stimulation signal according to the output parameter to stimulate the tongue muscle tissue or the hypoglossal nerve of the target object, so that the tongue muscle of the target object contracts and the airway is smooth, avoiding respiratory disorders caused by airway obstruction of the target object.

[0091] The target tissue can be a tongue muscle tissue of the target subject, such as the genioglossus muscle, the geniohyoid muscle, the palatoglossus muscle, and the like, or can be the hypoglossal nerve of the target subject. When the electrical stimulation device is an implanted electrical stimulation device, the electrode is attached to the tongue muscle of the target subject, so that the stimulation signal output by the electrical stimulation device can stimulate the tongue muscle tissue of the target subject. When the electrical stimulation device is an implanted electrical stimulation device, the electrode is built-in at the hypoglossal position of the target subject, so that the electrical stimulation device can stimulate the hypoglossal nerve of the target subject. The sensor can include a motion sensor, a temperature sensor, a pressure sensor, an impedance sensor, an oxygen sensor, and a posture sensor. The motion sensor collects motion data of the target subject during sleep, such as turning over, kicking legs, and the like. The temperature sensor can be arranged at the nasal cavity of the target subject or at the nostril of the target subject, and is used to collect the temperature of the exhaled gas when the target subject exhales. The pressure sensor can be arranged outside the chest cavity of the target subject or implanted in the chest of the target subject, and is used to collect the chest cavity pressure data of the target subject to determine the lung pressure of the target subject, so as to determine the breathing amplitude and breathing rhythm of the target subject. The impedance sensor can be arranged at the chest and abdomen of the target subject, and is used to determine the breathing rhythm of the target subject by collecting the electrical impedance change information of the chest and abdomen of the target subject. The oxygen sensor is used to collect the oxygen concentration in the body of the target subject. If the oxygen saturation in the body of the target subject is lower than a threshold value, it indicates that the target subject's breathing is blocked and the external oxygen cannot enter the lungs in time to supply the body of the target subject. The posture sensor can detect the continuous and significant changes of the body of the target subject, such as the change of the sleep posture, for example, the change from supine to lateral position. It should be noted that the breathing data at least includes all time sequences of the target subject in the current breathing cycle, or includes the breathing data of multiple breathing cycles in a historical time period, for example, the breathing data in the past ten minutes. The breathing cycle can be determined according to the breathing phase, from the 0-180° phase of the inspiration period to the 180-360° phase of the expiration period, which can be regarded as a complete breathing cycle.

[0092] After the sensor collects the breathing data of the target subject, the breathing data can be transmitted to the controller through the data transmission channel. After receiving the breathing signal, the controller can predict the predicted breathing state of the target subject in the next breathing cycle, and estimate the predicted sleep state of the target subject according to the predicted breathing state. Then, the predicted sleep state of the target subject is determined through the sleep breathing model combined with the physiological change information of the target subject in the sleep process. Then, according to the predicted sleep state, the time of outputting the stimulation signal, the time length of outputting the stimulation signal, the intensity of the stimulation signal, and the like are determined, and finally the electrode outputs the stimulation signal according to the output parameters to stimulate the muscle of the tongue of the target subject to keep the respiratory tract of the target subject unobstructed.

[0093] In some embodiments, based on the respiratory data, the process of determining the predicted sleep state of the target object can refer to the flowchart shown in FIG. 1, which includes the following steps: Figure 2

[0094] S201, according to the respiratory data, determining the predicted respiratory state of the target object in the next breathing cycle.

[0095] In some embodiments, after the sensor collects the respiratory data of the target object, the respiratory data can be transmitted to the controller through a data transmission channel. After receiving the respiratory signal, the controller can predict the predicted respiratory state of the target object in the next breathing cycle according to the respiratory state space model. The predicted respiratory state can include the motion state, respiratory frequency, phase, amplitude, blood oxygen content, and respiratory weakening frequency and phase of the target object. According to the predicted respiratory state of the target object, the risk of respiratory tract obstruction of the target object, the inspiratory duration, expiratory duration, and blood oxygen saturation change of the next breathing cycle can be determined.

[0096] In some embodiments, the current respiratory state of the target object can be determined according to the respiratory data first, and then the predicted respiratory state of the target object in the next breathing cycle can be determined according to the current respiratory state of the target object.

[0097] First, the respiratory state vector of the target object in the current breathing cycle can be constructed according to the respiratory data collected by the sensor, and then the respiratory state vector in the current breathing cycle is input into the preset respiratory state space model, and the respiratory state is predicted based on the preset respiratory state space model, so as to obtain the predicted respiratory state in the next breathing cycle. The respiratory state vector includes motion state, respiratory frequency, phase, amplitude, blood oxygen content, respiratory weakening frequency, and respiratory weakening phase. Among them, the state transition matrix in the respiratory state space model can be determined according to the respiratory state data sequence in the historical time period to estimate the optimal matrix. For example, it can be determined by maximum likelihood estimation, expectation maximization algorithm or least squares method.

[0098] In some embodiments, determining the respiratory state of the target object in the current breathing cycle can first determine a plurality of respiratory indicators of the target object, which can specifically include:

[0099] ​According to the thoracic pressure data, electrical impedance data and exhalation temperature data in the respiratory data, the current respiratory frequency, respiratory amplitude and respiratory phase of the target object are determined; according to the thoracic pressure data, electrical impedance data, blood oxygen data and exhalation temperature data in the respiratory data, the respiratory weakening frequency and respiratory weakening phase of the target object are determined; according to the blood oxygen data in the respiratory data, the current blood oxygen content of the target object is determined; and according to the motion data and posture data in the respiratory data, the motion state of the target object is determined.

[0100] After obtaining the respiratory data collected by the sensors, the respiratory data can be preprocessed first, for example, the respiratory data can be synchronously sampled to ensure that all sensor data time stamps are aligned (using hardware synchronization or software interpolation), the respiratory data can also be processed, and different sensor signals can also be normalized to eliminate the influence of dimension.

[0101] During different stages of the respiratory process, the thoracic pressure, thoracic electrical impedance and exhalation temperature all change with the respiratory process. For example, the thoracic pressure increases in negative pressure during inhalation and recovers during exhalation; the thoracic electrical impedance increases during inhalation (air entering reduces conductivity) and decreases during exhalation; the temperature during exhalation is significantly higher than that during inhalation (exhaled gas ≈ body temperature, inhaled gas ≈ room temperature). Therefore, the current respiratory frequency, respiratory amplitude and respiratory phase of the target object can be determined based on the thoracic pressure data, electrical impedance data and exhalation temperature data in the respiratory data. For example, according to the electrical impedance data during the respiratory process, the characteristics of the electrical impedance data are extracted by frequency spectrum analysis method, so as to determine the respiratory frequency, or the peak value test method can also be used to analyze the minimum values in the pressure data and exhalation temperature, and the interval between adjacent minimum values is determined to determine the respiratory period, and further determine the respiratory frequency, and further, multi-sensor cross verification can be used to compare the frequencies calculated by the three signals, and the abnormal value is proposed, so as to determine the respiratory frequency.

[0102] It should be noted that in the sleep state, the target object may cause the respiratory signal collected by the motion sensor to be weak due to body position change, and at the same time cause the environmental noise in the respiratory signal to be seriously disturbed, for example Figure 2 As shown in the schematic diagram, the respiratory signal includes obvious noise interference, causing the respiratory cycle boundary in the respiratory signal to be blurred. In this case, it will further cause the estimation accuracy of the respiratory frequency and the respiratory phase to be low, and if the estimation accuracy of the respiratory frequency and the respiratory phase is low, the accuracy of the electrical stimulation parameters determined according to the current respiratory state will be low, which will further cause the stimulation accuracy of the hypoglossal nerve or the tongue muscle tissue of the target object to be low, and the hypoglossal nerve or the tongue muscle tissue of the target object cannot be accurately stimulated.

[0103] Based on this, in some embodiments, feature extraction can be performed on the respiratory signal carrying the interference signal to improve the accuracy of determining the respiratory frequency and the respiratory phase. For details, please refer to the flowchart shown in FIG. 3, which includes the following steps: Figure 3

[0104] S301, in response to determining that the respiratory signal of the target object contains an interference signal, determining a sampling window according to a preset respiratory frequency range and a preset respiratory motion feature.

[0105] S302, performing feature extraction on the respiratory signal in the sampling window according to the preset respiratory motion feature to determine a feature weight.

[0106] S303, determining a simulated respiratory signal according to a reference respiratory frequency and a reference respiratory phase.

[0107] S304, adjusting the reference respiratory frequency and the reference respiratory phase according to the feature weight to minimize the residual error between the respiratory signal and the simulated respiratory signal, and determining the current respiratory frequency and the current respiratory phase in the current respiratory state.

[0108] The preset respiratory frequency range can be determined according to the historical respiratory data of the target object, and the preset respiratory motion feature can be determined by performing feature extraction on the historical respiratory data of the target object. The respiratory signal can be determined according to the motion data in the respiratory data, for example, the respiratory signal can be determined according to the chest displacement data of the target object. According to the respiratory feature in the respiratory signal, it can be determined whether there is an interference signal in the respiratory signal. For example, when there are data points in the respiratory signal that exceed the preset amplitude range, it can be determined that there is an interference signal in the respiratory signal. Or, if the change rate (slope) between adjacent sampling points in the respiratory signal exceeds the preset threshold, it can be determined that there is an interference signal in the respiratory signal.

[0109] In the case where it is determined that the respiratory signal of the target object contains an interference signal, the length of the sampling window can be determined according to the preset respiratory frequency range and the preset respiratory motion feature, so that the sampling window contains a certain number of respiratory cycles. For example, if the preset respiratory frequency range is 3-6 times per minute, the length of the sampling window can be determined as 100s, so that the sampling window contains at least 5 respiratory cycles. For another example, according to the preset respiratory motion feature, the sampling window in the respiratory signal contains a selected smooth, periodic and sinusoidal motion. By selecting the sampling window, the calculation resources can be saved in the invalid frequency band, the interference signal in the sampling window can be preliminarily reduced, and the subsequent calculation efficiency can be improved.

[0110] ​After the sampling window is determined, the respiratory signal in the sampling window can be feature extracted according to the preset respiratory motion feature to determine a feature weight. Specifically, a filter can be designed according to the preset respiratory motion feature, and the respiratory signal in the sampling window can be convoluted by the filter to enhance the signal component (i.e., the respiratory signal) similar to the preset respiratory motion feature in the sampling window and weaken the dissimilar component (i.e., the noise), thereby improving the signal-to-noise ratio of the respiratory signal in the sampling window. After the respiratory signal in the sampling window is convoluted, a weight feature vector of the respiratory signal in the sampling window can be obtained, and the feature weight represents the weight of different respiratory frequencies.

[0111] The reference respiratory frequency can be determined according to the preset respiratory frequency range, for example, by selecting a respiratory frequency in the preset respiratory frequency range. The reference respiratory frequency can also be determined according to the feature weight, for example, by determining the respiratory frequency corresponding to the maximum weight value in the feature weight as the reference respiratory frequency. After the respiratory frequency is determined, a corresponding reference respiratory phase can be selected according to the respiratory frequency, and the reference respiratory phase can also be determined according to the preset respiratory motion feature. After the reference respiratory frequency and the reference respiratory phase are determined, a respiratory model can be constructed according to the reference respiratory frequency and the reference respiratory phase, and an analog respiratory signal can be obtained according to the respiratory model. The respiratory model can be a sine wave function, and the analog respiratory signal can be obtained by substituting the reference respiratory frequency and the reference respiratory phase into the sine wave function.

[0112] The reference respiratory frequency and the reference respiratory phase are adjusted according to the feature weight to minimize the residual error between the respiratory signal and the analog respiratory signal, and the current respiratory frequency and the current respiratory phase in the current respiratory state are determined.

[0113] Then, the residual error between the respiratory signal and the analog respiratory signal is calculated, and a cost function is constructed according to the feature weight and the residual error between the respiratory signal and the analog respiratory signal. The values of the reference respiratory frequency and the reference respiratory phase are iteratively adjusted by minimizing the cost function until the cost function converges, and the reference respiratory frequency at this time is determined as the current respiratory frequency in the current respiratory state, and the reference respiratory phase at this time is determined as the current respiratory phase in the current respiratory state. In some embodiments, the residual error between the respiratory signal and the analog respiratory signal can be minimized by algorithms such as Gradient Descent, Gauss-Newton, or Levenberg-Marquardt to obtain the current respiratory frequency and the current respiratory phase.

[0114] In determining the respiratory amplitude, the respiratory amplitude can be determined according to the weighted calculation of the peak value in the electrical impedance data and the peak value in the thoracic pressure data, and the exhaled air volume can be determined according to the integral area of the respiratory temperature data, so as to determine the respiratory amplitude. In determining the respiratory phase, the respiratory cycle can be segmented based on a pre-constructed respiratory cycle template, for example, the rising edge (exhalation start) of the respiratory temperature data is used as the cycle starting point, and then the key point detection is performed, for example, the end of inspiration point is determined according to the peak value time of the thoracic pressure data and / or the electrical impedance data, and the end of exhalation point is determined according to the time when the exhalation temperature data falls back to the baseline, so as to determine the respiratory phase.

[0115] According to the thoracic pressure data, the electrical impedance data, the blood oxygen data and the exhalation temperature data in the respiratory data, the respiratory weakening frequency and the respiratory weakening phase of the target object are determined. The respiratory weakening frequency refers to the frequency of abnormal breathing patterns of the target object caused by repeated collapse of the upper airway during sleep, including the apnea frequency and / or the hypopnea frequency. The apnea frequency refers to the time when the respiratory airflow is completely stopped and exceeds a threshold, but the thoraco-abdominal breathing movement still exists (caused by obstruction). The apnea frequency can be determined according to the number of apneas per hour of sleep, for example, if the target object has 30 apneas per hour, the apnea frequency is 30 times per hour. The specific respiratory data collected by the sensor is that the exhalation temperature data disappears (indicating that there is no airflow in the nasal cavity), but the thoracic pressure data and / or the electrical impedance data indicate that the target object is in a breathing state, such as a reduced thoracic pressure fluctuation amplitude (reduced inspiration effort) / reduced electrical impedance change amplitude (insufficient chest expansion). At the same time, the blood oxygen data indicates that the blood oxygen saturation in the target object decreases. The hypopnea frequency refers to the partial reduction of respiratory airflow and lasts more than a threshold, accompanied by a decrease in blood oxygen saturation or micro-awakening. The hypopnea frequency can be determined according to the number of hypopneas per hour of sleep, for example, if there are 20 hypopneas per hour, the hypopnea frequency is 20 times per hour. The specific respiratory data collected by the sensor is that the exhalation temperature peak value is reduced or the plateau period is shortened (airflow is reduced), but the thoracic pressure fluctuation amplitude is reduced (inspiration effort is reduced) / electrical impedance change amplitude is reduced (chest expansion is insufficient), and at the same time, the blood oxygen data indicates that the blood oxygen saturation in the target object decreases.

[0116] The respiratory weakening phase refers to a specific stage of reduced ventilation efficiency in a single respiratory cycle due to partial or complete collapse of the upper airway, mainly including inspiratory obstruction, i.e., the upper airway collapses during inspiration, airflow is limited but respiratory effort exists, and expiratory limitation, i.e., airflow is prolonged due to airway narrowing or dynamic collapse during expiration. The detection method of the inspiratory weakening phase needs to perform cycle segmentation and feature extraction steps. First, the temperature rise of the exhalation is used to mark the beginning of the exhalation (cycle start point), and the temperature falls from the lowest point to the peak value of the electrical impedance / pressure as the inspiratory phase. Then, according to the chest pressure data features, electrical impedance data features, and blood oxygen data features, the inspiratory weakening data is determined, wherein the chest pressure data features include enhanced negative pressure fluctuations (sawtooth waveforms reflecting efforts to resist obstruction), the electrical impedance data features include a slow impedance rise speed (due to limited airflow, limited chest expansion), and the blood oxygen data features include a decrease in blood oxygen saturation. The expiratory weakening phase needs to first locate the expiratory phase according to the time of the peak value of the chest pressure data, the time of the peak value of the electrical impedance data, or the time of the exhalation temperature data falling to the baseline, and then determine the expiratory weakening phase according to the exhalation temperature data plateau, the chest pressure exhalation not returning to the baseline, and other data features.

[0117] And the current blood oxygen content of the target object can be determined according to the blood oxygen data in the respiratory data. And the motion state of the target object can be determined according to the motion data and the posture data in the respiratory data. The motion state represents the body movement or posture of the target object during sleep. The number of times of turning over, the acceleration standard deviation (reflecting the amplitude of movement) can be determined according to the time domain features in the motion data. The frequency domain features can be extracted according to the Fourier transform to determine the movement of the target object, such as the typical frequency of 0.1-0.3 Hz of the turning over movement. The continuity of the movement can also be determined according to the periodicity features. The spatial relationship of the body movement of the target object can also be extracted according to the posture data, and then the sleep posture of the target object can be determined, for example, the prone position can be identified according to the included angle between the midpoint of the shoulder and the key point of the nose (the included angle is less than a critical value).

[0118] After obtaining the current respiratory state of the target object, the controller can predict the predicted respiratory state of the target object in the next respiratory cycle based on the current respiratory state of the target object. The method for determining the predicted respiratory state is shown in the flowchart of Figure 4 The flowchart includes:

[0119] S401, feature extraction is performed on the current respiratory state to obtain a feature vector.

[0120] S402, based on the state vector corresponding to the current respiratory state and the feature vector, the predicted respiratory state of the target object in the next respiratory cycle is predicted through a pre-constructed respiratory state space model.

[0121] The state transition matrix in the respiratory state space model can be determined based on a respiratory state data sequence in a historical time period. When performing feature extraction on the current respiratory state, the convolutional feature extraction network can be used to extract the feature vector corresponding to the current respiratory state. The current respiratory state is extracted to obtain a feature vector corresponding to the current respiratory state. Then, the current respiratory state and the feature vector are input into a pre-constructed respiratory state space model to predict the respiratory state of the next respiratory cycle. In an embodiment, the calculation method of the predicted respiratory state can refer to formula (1):

[0122]

[0123] (1)

[0124] wherein, represents the predicted respiratory state, represents the current respiratory state, represents the state transition matrix, represents the feature vector, and G is a weight matrix.

[0125] The weight matrix is used to map the feature vector to the state space, and the elements thereof reflect the contribution strength of different features to the next cycle state. The weight matrix can explicitly quantify the influence of the feature on the state, and can include prior information related to the respiratory cycle, amplitude, blood oxygen content, etc., such as that a prolonged respiratory cycle corresponds to a reduced frequency, a decreased respiratory amplitude corresponds to a reduced blood oxygen saturation, a rapid decrease in blood oxygen saturation corresponds to a reduced respiratory frequency, a current reduced respiratory frequency corresponds to a continuously weakened next respiratory cycle, and the like.

[0126] After obtaining the predicted respiratory state of the target object in the next respiratory cycle, the controller can further determine physiological change data of the target object during sleep, such as sleep posture change data, physiological index change data, etc.

[0127] S202, according to the posture data and the electrical impedance data in the respiratory data, determining a sleep posture signal of the target object.

[0128] The sleep position signal data of the target object refers to the signal of changes in body posture (such as supine, lateral, prone, etc.) during sleep. When determining the sleep position signal of the target object, the changes in the target object's body values ​​along the X, Y, and Z axes and the electrical impedance data can be extracted based on the posture data detected by the accelerometer. These changes are then combined with preset judgment conditions for position changes to determine the target object's sleep position signal. For example, the target object's sleep posture can be determined based on the static gravity component in the posture data, where supine corresponds to Z-axis ≈ 1g, X / Y ≈ 0g; left lateral position corresponds to X-axis ≈ -1g, Y / Z ≈ 0g; and prone corresponds to Z-axis ≈ -1g. Alternatively, the target object's position change can be determined through dynamic roll detection, such as the peak value of the gyroscope's Y-axis angular velocity (typically a rolling angular velocity > 50° / s). A decrease in abdominal impedance in the electrical impedance data (due to abdominal pressure and improved electrode contact) can characterize a change in the target object from supine to lateral position.

[0129] S203, determine the sleep movement signal of the target object based on the motion data and thoracic pressure data in the respiratory data.

[0130] The sleep motion signals refer to the body movement signals caused by limb movements, turning over, or brief micro-awakenings that occur during sleep. The timing of the start and end of movement can be determined based on motion data. Combining this with intrathoracic pressure data can further determine whether the target's respiratory status has changed. For example, if the pressure value reaches a peak after the motion data indicates the start of movement, it indicates that the target's limbs are moving. Conversely, if the intrathoracic pressure data changes but the motion data indicates no limb movement, it can be determined that the target may have simply coughed or the motion sensor may have shifted.

[0131] S204, based on the blood oxygen data in the respiratory data, the predicted respiratory state, the sleep position signal, and the sleep movement signal, determine the predicted sleep state of the target object.

[0132] In determining the predicted sleep state of the target object according to the blood oxygen data, the predicted respiratory state, the sleep position signal and the sleep motion signal, a sleep respiratory model can be used for determination, which can be a hierarchical mixed model. First, feature extraction is performed on the blood oxygen data, the predicted respiratory state, the sleep position signal and the sleep motion signal. Specifically, frequency, amplitude, inspiration slope and other features in the respiratory signal can be extracted; supine / lying classification, turning frequency and other features in the sleep position signal can be extracted; duration features in the sleep motion signal can be extracted; and mean oxygen saturation, descent slope, minimum value and other features in the blood oxygen data can be extracted. Then, according to the feature information of the predicted respiratory state, a hidden Markov model is used to calculate the probability of being in different sleep stages, which can include NREM (Non-Rapid Eye Movement Sleep, non-rapid eye movement sleep), REM (Rapid Eye Movement Sleep, rapid eye movement sleep) and wakefulness. And the weight parameters are generated according to the features of the sleep position signal and the sleep motion signal to correct the obtained probability of different sleep stages; and the features extracted from the blood oxygen data are used to verify the sleep respiratory stage. In an embodiment, the predicted sleep state of the target object can be determined by referring to the method shown in formula (2), as follows:

[0133] (2)

[0134] wherein, represents the predicted sleep state, represents the sleep respiratory model, represents the predicted respiratory state, represents the sleep motion signal, and represents the blood oxygen content.

[0135] After obtaining the predicted sleep state of the target object, the output parameters of the stimulation signal can be determined according to the predicted sleep state. The output parameters of the stimulation signal include whether to output the stimulation signal, the output time of the stimulation signal, the output duration of the stimulation signal and the intensity of the stimulation signal.

[0136] In some embodiments, according to the predicted sleep state, the output parameters of the stimulation signal are determined, including: in response to the predicted sleep state representing that the target object starts to enter sleep, determining the output parameters of the stimulation signal according to the sleep data of the target object in a historical time window; and in response to the predicted sleep state representing that the target object is in a wakeful state, stopping outputting the stimulation signal.

[0137] The target object begins to enter sleep when its predicted sleep state changes from wakefulness to NREM or REM. Therefore, the output parameters of the stimulus signal can be determined based on the target object's predicted sleep state within the historical event window. For example, when the target object begins to enter sleep, the output parameters of the stimulus signal can be determined based on the target object's sleep data over the past ten minutes. Specifically, the output parameters can be determined based on the respiratory status data within the sleep data. In one embodiment, the next respiratory cycle can be determined first, as detailed in formula (3):

[0138] (3)

[0139] in, This is a respiratory state cycle, through the state... Each dimension The weighted average calculation uses the weight of the k-th dimension. This indicates that the next respiratory cycle can be obtained.

[0140] In one embodiment, the stimulus output timing can be determined based on a preset time bias and the predicted start time of respiration in the respiratory state. It should be noted that the stimulus signal can be emitted before the start of the next respiratory cycle. The preset time bias can be determined based on nerve conduction delay and muscle mechanical response delay, typically ranging from 200ms to 500ms, and can be calibrated according to different target subjects. Based on the respiratory state cycle determined in the predicted respiratory state, the start time of the next respiratory cycle can be determined. Then, before the start time of the next respiratory cycle, the stimulus output timing is determined based on the preset time bias.

[0141] The dynamic time bandwidth refers to the time domain window corresponding to the frequency range of effective physiological changes in the respiratory signal (e.g., 0.1-2 Hz corresponds to 0.5-10 seconds), which can be determined by the power spectral density (PSD) of the respiratory signal. The predicted duration of the respiratory state is the next respiratory state cycle. After determining the next respiratory state cycle, the stimulation duration can be determined according to the method shown in formula (4), as follows:

[0142] (4)

[0143] Where h is the stimulation duration, For dynamic time bandwidth.

[0144] When determining the intensity of the stimulus signal, the intensity of the stimulus signal can be obtained by fitting the respiratory amplitude and blood oxygen content with a polynomial. For details, please refer to formula (5):

[0145] (5)

[0146] wherein, wherein , is a polynomial coefficient, n is the degree of the polynomial, is a breathing amplitude, is a blood oxygen content.

[0147] After obtaining the output parameter of the stimulation signal for stimulating the tongue muscle tissue and hypoglossal nerve, the controller can control the electrode to output the stimulation signal according to the output parameter to electrically stimulate the hypoglossal nerve or the tongue muscle tissue of the target object, so as to ensure that the stimulation signal can accurately stimulate the tongue muscle tissue or the hypoglossal nerve of the user, keep the airway of the user unobstructed, and improve the sleep experience of the user.

[0148] If the predicted sleep state indicates that the target object is in a wake state, it means that the target object can self-control the contraction of the tongue muscle without the need of the electric stimulation device to additionally output the stimulation signal. Therefore, the controller can control the electrode to stop outputting the stimulation signal, so as to avoid the influence of the stimulation signal on the falling asleep of the target object, shorten the falling asleep time of the target object, and improve the sleep quality of the target object.

[0149] In some embodiments, the controller is further configured to: determine a correction function of the respiratory state space model based on a difference between a real-time respiratory state of the target object in a next breathing cycle and the predicted respiratory state; and correct the respiratory state space model based on the correction function.

[0150] In an embodiment, the correction function can refer to formula (6):

[0151] (6)

[0152] wherein, represents a predicted sleep state, represents an actual predicted sleep state. After obtaining the correction function , the correction function can be superimposed on the sleep respiratory model to correct the sleep respiratory model.

[0153] The stimulation signal is generated by fusing the sleep respiratory state, and the adaptability of the stimulation signal in the sleep process is optimized by combining the sleep respiratory state feedback in the stimulation process. When the target object has events such as awakening and getting up during the sleep process, the electric stimulation device adaptively adjusts the stimulation signal according to the change of the respiratory state, avoids the influence on the sleep process of the target object, and improves the adaptability of the stimulation of obstructive sleep apnea.

[0154] The disclosure also provides a method for determining an electric stimulation signal, which can be applied to an electric stimulation device. The method steps are described in detail in Figure 5including:

[0155] S501, acquiring, by a sensor, breathing data of a target object, wherein the breathing data comprises at least one of motion data, thoracic pressure data, electrical impedance data, blood oxygen data, and posture data.

[0156] S502, determining, by a controller, a predicted sleep state of the target object based on the breathing data.

[0157] S503, determining, by the controller, an output parameter of a stimulation signal based on the predicted sleep state.

[0158] S504, controlling, by the controller, an electrode to output the stimulation signal based on the output parameter of the stimulation signal.

[0159] The sensor collects breathing data of the target object during sleep and transmits the breathing data to the controller. The controller predicts the predicted sleep state of the target object according to the breathing data collected by the sensor, and then determines the output parameter of the stimulation signal for neuromuscular stimulation of the target object according to the predicted sleep state of the target object. Finally, the electrode outputs the stimulation signal according to the output parameter to stimulate the tongue muscle tissue or hypoglossal nerve of the target object, so that the tongue muscle of the target object contracts and keeps the respiratory tract smooth, avoiding respiratory disorders caused by respiratory tract obstruction.

[0160] In some embodiments, the determining, based on the breathing data, of the predicted sleep state of the target object comprises:

[0161] determining, based on the breathing data, a predicted breathing state of the target object in a next breathing cycle;

[0162] determining, based on posture data and electrical impedance data in the breathing data, a sleep position signal of the target object;

[0163] determining, based on motion data and thoracic pressure data in the breathing data, a sleep movement signal of the target object;

[0164] determining, based on blood oxygen data in the breathing data, the predicted breathing state, the sleep position signal, and the sleep movement signal, the predicted sleep state of the target object.

[0165] The determining, based on the breathing data, of the predicted breathing state of the target object in a next breathing cycle comprises:

[0166] determining, based on the breathing data, a current breathing state of the target object;

[0167] determining, based on the current breathing state of the target object, the predicted breathing state of the target object in a next breathing cycle.

[0168] In some embodiments, the determining the current respiratory state of the target object according to the respiratory data comprises:

[0169] In response to determining that there is an interference signal in the respiratory signal of the target object, determining a sampling window according to a preset respiratory frequency range and a preset respiratory motion feature, wherein the respiratory signal is determined according to motion data in the respiratory data;

[0170] extracting a feature weight of the respiratory signal in the sampling window according to the preset respiratory motion feature;

[0171] determining a simulated respiratory signal according to a reference respiratory frequency and a reference respiratory phase;

[0172] adjusting the reference respiratory frequency and the reference respiratory phase according to the feature weight to minimize a residual error between the respiratory signal and the simulated respiratory signal, and determining a current respiratory frequency and a current respiratory phase in the current respiratory state.

[0173] In some embodiments, the determining the predicted respiratory state of the target object in the next respiratory cycle according to the current respiratory state of the target object comprises:

[0174] based on a pre-constructed respiratory state space model;

[0175] extracting a feature vector from the current respiratory state;

[0176] predicting the predicted respiratory state of the target object in the next respiratory cycle through the state space model based on a state vector corresponding to the current respiratory state and the feature vector.

[0177] In some embodiments, the determining the output parameter of the stimulation signal according to the predicted sleep state comprises:

[0178] In response to the predicted sleep state representing that the target object starts to enter sleep, determining the output parameter of the stimulation signal according to sleep data of the target object in a historical time window;

[0179] In response to the predicted sleep state representing that the target object is in a wake state, stopping outputting the stimulation signal.

[0180] In some embodiments, the method further comprises:

[0181] determining a correction function of the respiratory state space model based on a difference between a real-time respiratory state in the next respiratory cycle of the target object and the predicted respiratory state;

[0182] correct the respiratory state space model based on the correction function.

[0183] In some embodiments, the controller is further configured to:

[0184] update the respiratory state space model based on a difference between the actual respiratory state and the predicted respiratory state in a next respiratory cycle of the target object;

[0185] adjust a collection parameter of the sensor based on the predicted respiratory state, wherein the collection parameter of the sensor comprises at least one of a collection frequency, a resolution, and a power.

[0186] In some embodiments, the method for determining the electrical stimulation signal can refer to the flowchart shown in FIG. 6. Figure 6

[0187] First, the controller controls a plurality of sensors to collect respiratory data of a target object when wearing an electrical stimulation device through step 1. Then, the respiratory data is analyzed through step 2 to determine the sleep position, sleep body movement, and blood oxygen content of the target object. Further, a sleep respiratory model is called through step 3 to estimate the predicted sleep state of the target object according to the sleep position, sleep body movement, and blood oxygen content of the target object, so as to determine the output parameter of the electrical stimulation device according to the predicted sleep state of the target object in step 4, and control the electrode to output the stimulation signal to stimulate the hypoglossal nerve or the tongue muscle tissue of the target object. In step 5, the collection parameter of the sensor is updated according to the estimated predicted sleep state and the actual predicted sleep state of the target object, and in step 6, the sleep respiratory model is updated according to the estimated predicted sleep state and the actual predicted sleep state of the target object to improve the accuracy of the sleep respiratory model in predicting the predicted sleep state of the target object.

[0188] For each of the method embodiments described above, in order to simply describe, it is expressed as a series of action combinations, but those skilled in the art should know that the disclosure is not limited by the order of the described actions, because according to the disclosure, certain steps can be performed in other orders or simultaneously.

[0189] Secondly, those skilled in the art should know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the disclosure.

[0190] ​The above describes particular embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, acts or steps recited in the claims can be performed in a different order and still accomplish the desired results. Additionally, the processes depicted in the figures do not necessarily require the particular order shown or sequential order to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous or necessary.

[0191] Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the features of the disclosure as set forth herein. It is intended that the disclosure be construed as including any paterns of embodiments and their equivalents following the principles of the disclosure and including modifications and improvements to the paterns of the disclosure as would occur to those skilled in the art to which the disclosure pertains. The specification and examples given herein are intended as illustrative only and not as limiting. The true scope of the disclosure is set forth in the following claims.

[0192] It is to be understood that the present disclosure is not limited to the precise construction described above and shown in the attached drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the present disclosure. The scope of the present disclosure is limited only by the claims that follow.

[0193] The above description is merely the preferred embodiments of the present disclosure and is not intended in any way to limit the present disclosure. Any modification, equivalent replacement or improvement made without departing from the spirit and principle of the present disclosure shall be included in the scope of the present disclosure.

Claims

1. An electrical stimulation device, characterized in that, The electrical stimulation device includes: An electrode that, in use, comes into contact with a target object and is used to electrically stimulate target tissues in the target object's body. A sensor for collecting respiratory data of the target object, the respiratory data including at least one of the following: motion data, expiratory temperature data, intrathoracic pressure data, electrical impedance data, blood oxygen data, and posture data; A controller, electrically connected to the sensor and the electrode, is used for: Based on the respiratory data collected by the sensor, the predicted sleep state of the target object is determined; Based on the predicted sleep state, the output parameters of the stimulation signal are determined; Based on the output parameters of the stimulation signal, the electrode is controlled to output a stimulation signal to stimulate the target tissue; Determining the predicted sleep state of the target subject based on the respiratory data collected by the sensor includes: Based on the respiratory data, the predicted respiratory state of the target object in the next respiratory cycle is determined; Based on the blood oxygen data in the respiratory data, the predicted respiratory state, sleep posture signals and sleep movement signals, the predicted sleep state of the target object is determined by a sleep breathing model. Determining the predicted respiratory state of the target object in the next respiratory cycle based on the respiratory data includes: Based on the respiratory data, determine the current respiratory state of the target object; Based on the target object's current respiratory state, determine the predicted respiratory state of the target object in the next respiratory cycle.

2. The electrical stimulation device according to claim 1, characterized in that, Determining the predicted sleep state of the target subject based on the respiratory data collected by the sensor includes: Based on the posture data and electrical impedance data in the respiratory data, the sleep position signal of the target object is determined; Based on the motion data and thoracic pressure data in the respiratory data, the sleep movement signals of the target subject are determined.

3. The electrical stimulation device according to claim 1, characterized in that, Determining the current respiratory state of the target object based on the respiratory data includes: In response to the determination that there is an interference signal in the respiratory signal of the target object, a sampling window is determined according to a preset respiratory frequency range and preset respiratory motion characteristics, wherein the respiratory signal is determined based on the motion data in the respiratory data; Based on the preset respiratory motion characteristics, feature extraction is performed on the respiratory signal within the sampling window, and feature weights are determined. The simulated respiratory signal is determined based on the reference respiratory rate and reference respiratory phase; The reference respiratory rate and reference respiratory phase are adjusted according to the feature weights to minimize the residual between the respiratory signal and the simulated respiratory signal, and to determine the current respiratory rate and current respiratory phase in the current respiratory state.

4. The electrical stimulation device according to claim 1, characterized in that, The step of determining the predicted respiratory state of the target object in the next respiratory cycle based on the target object's current respiratory state includes: Feature vectors are obtained by extracting features from the current breathing state. Based on the state vector corresponding to the current breathing state and the feature vector, the predicted breathing state of the target object in the next breathing cycle is predicted through a pre-constructed breathing state space model.

5. The electrical stimulation device according to claim 1, characterized in that, The step of determining the output parameters of the stimulation signal based on the predicted sleep state includes: In response to the predicted sleep state indicating that the target object has begun to enter sleep, the output parameters of the stimulation signal are determined based on the sleep data of the target object within a historical time window. In response to the predicted sleep state indicating that the target object is awake, the output of the stimulation signal is stopped.

6. The electrical stimulation device according to claim 2, characterized in that, The controller is also used for: Based on the difference between the actual sleep state and the predicted sleep state of the target object, a correction function for the sleep breathing model is determined. The sleep breathing model is modified based on the aforementioned correction function.

7. The electrical stimulation device according to claim 4, characterized in that, The controller is also used for: The respiratory state space model is updated based on the difference between the actual respiratory state and the predicted respiratory state of the target object in the next respiratory cycle. Based on the predicted respiratory state, the acquisition parameters of the sensor are adjusted, wherein the acquisition parameters of the sensor include at least one of the following: acquisition frequency, resolution, and power.

Images