Intelligent sleep-aiding earphone based on electrocardio monitoring

Abstract

The present application relates to sleep-aiding earphone technical field, especially to a kind of intelligent sleep-aiding earphone based on electrocardiogram monitoring, obtains original electrocardiogram signal and carries out detection analysis to obtain RR interval sequence and R wave amplitude sequence, based on RR interval sequence carries out heart rate variability analysis to obtain RR interval standard deviation, first signal frequency and second signal frequency, and then determine corresponding sleep stage, based on sleep stage to play corresponding audio to intervene the sleep process of wearer, based on R wave amplitude sequence analysis obtains R wave amplitude variation coefficient, and then determine whether audio intervention process is qualified, and then based on the determination result selects to maintain current audio playback or the rhythm frequency of audio is dynamically adjusted to revise audio intervention process.The present application can improve the consistency of the electrocardiogram frequency of wearer and the audio frequency of earphone, so as to realize sleep-aiding.

Description

Technical Field

[0001] This invention relates to the field of sleep aid headphones, and more particularly to a smart sleep aid headphone based on electrocardiogram monitoring. Background Technology

[0002] Sleep disorders have become a major public health problem affecting the physical and mental health of hundreds of millions of people worldwide. Insomnia, difficulty falling asleep, and poor sleep quality not only lead to daytime fatigue, decreased attention, and mood swings, but are also significantly associated with a long-term increased risk of various chronic diseases such as cardiovascular disease, metabolic disorders, and weakened immunity. Therefore, there is an urgent need for a smart headset that can help insomniacs, intelligently selecting and adjusting various audio intervention strategies based on real-time analysis of physiological states.

[0003] Chinese patent application publication number CN214679630U discloses a sleep headset based on electrocardiogram (ECG) monitoring. This technical solution uses Bluetooth headphones to first play sleep-aid music. When the patient's ECG data gradually decreases to the sleep-state heart rate, the volume is reduced while the sleep-aid music continues to play. When the patient's ECG signal is detected to be within a set range for normal sleep-state heart rate, the sleep-aid music is turned off, thereby helping the patient quickly fall asleep. However, this related technical solution only controls music playback based on a static heart rate threshold when analyzing ECG signals, failing to consider the differences in sensitivity to sound stimulation at different sleep stages and the simplistic approach to sleep assessment and intervention. Summary of the Invention

[0004] To address this issue, the present invention provides a smart sleep aid headphone based on electrocardiogram (ECG) monitoring. This overcomes the problem that existing technologies only control the playback of sleep aid music based on static heart rate thresholds, without comprehensively considering the differences in sensitivity to sound stimulation at different sleep stages, and with a single sleep judgment and intervention method. Consequently, they fail to improve the consistency between the sleep aid audio frequency and the wearer's ECG signal frequency, thus affecting the sleep aid efficiency.

[0005] To achieve the above objectives, the present invention provides a smart sleep aid headphone based on electrocardiogram monitoring, comprising:

[0006] The ECG acquisition module is used to acquire the wearer's raw ECG signal in real time and preprocess it to obtain the processed ECG signal;

[0007] The R-wave detection module is connected to the ECG acquisition module and is used to perform R-wave detection and analysis based on the processed ECG signal to obtain the RR interval sequence and the R-wave amplitude sequence.

[0008] The feature extraction module, which is connected to the R-wave detection module, is used to perform heart rate variability analysis based on the RR interval sequence to obtain the RR interval standard deviation, the first signal frequency and the second signal frequency, and to perform analysis based on the R-wave amplitude sequence to obtain the R-wave amplitude variation coefficient.

[0009] A sleep stage module, which is connected to the feature extraction module, is used to determine the corresponding sleep stage based on the standard deviation of the RR interval, the first signal frequency, and the second signal frequency.

[0010] An audio playback module, which is connected to the sleep stage, is used to pre-store several different types of audio and determine the basic value of the electrocardiogram frequency based on the sleep stage, so as to play the corresponding type of audio to perform audio intervention on the wearer.

[0011] An analysis module, which is connected to the feature extraction module and the audio playback module respectively, is used to determine whether the audio intervention process is qualified based on the changes in the R-wave amplitude variation coefficient and the respiratory wave harmonic-to-noise ratio.

[0012] The intelligent control module is connected to both the analysis module and the audio playback module, and is used to maintain the current playback of the audio or dynamically adjust the rhythm and frequency of the audio based on the results determined by the audio intervention process.

[0013] Furthermore, the sleep stage module is also used to determine the corresponding sleep stage based on the comparison results of the standard deviation of the RR interval with the preset standard deviation of the RR interval and the comparison results of the frequency ratio with the preset frequency ratio;

[0014] The sleep stages include the sleep onset stage, light sleep stage, and deep sleep stage, and the frequency ratio is the ratio of the first signal frequency to the second signal frequency.

[0015] Furthermore, the audio playback module is also used to play audio of the type corresponding to the current sleep stage, wherein different types of audio correspond to different rhythm frequencies and correspond one-to-one with sleep stages.

[0016] Furthermore, the analysis module is also used to determine whether the audio interference process is qualified based on the comparison results of the R-wave amplitude variation coefficient with the preset R-wave amplitude variation coefficient and the comparison results of the respiratory wave harmonic noise ratio with the critical respiratory wave harmonic noise ratio.

[0017] If the R-wave amplitude variation coefficient is greater than the first preset R-wave amplitude variation coefficient and less than or equal to the second preset R-wave amplitude variation coefficient, the audio intervention process is re-determined based on the comparison result of the respiratory wave harmonic noise ratio and the critical respiratory wave harmonic noise ratio.

[0018] Furthermore, the analysis module is also used to adjust the rhythm frequency of the currently played audio based on the comparison result between the respiratory wave harmonic ratio and the critical respiratory wave harmonic ratio.

[0019] If the respiratory wave harmonic noise ratio is less than or equal to the critical respiratory wave harmonic noise ratio, it is determined that there is an abnormality in the audio intervention process, and the rhythm frequency of the audio is adjusted based on the harmonic noise ratio difference.

[0020] The harmonic noise ratio difference is the difference between the critical respiratory wave harmonic noise ratio and the respiratory wave harmonic noise ratio.

[0021] Furthermore, the analysis module is also used to determine the reduction of the rhythm frequency of the audio based on the comparison result between the harmonic-to-noise ratio difference and the preset harmonic-to-noise ratio difference, and the reduction of the audio rhythm frequency is positively correlated with the harmonic-to-noise ratio difference.

[0022] Furthermore, the analysis module is also used to determine, within a preset duration after reducing the rhythm frequency of the audio, the wearer's current sleep process is in an abnormal and disordered state based on the comparison result that the harmonic noise ratio difference is greater than the second preset harmonic noise ratio difference.

[0023] The intelligent control module is also used to emit interfering audio to interrupt the wearer's sleep state, wherein the preset harmonic noise ratio difference includes a second preset harmonic noise ratio difference.

[0024] Furthermore, the analysis module is also used to determine that the audio intervention process is unqualified within a preset duration after reducing the rhythm frequency of the audio, based on the comparison result that the respiratory wave harmonic ratio is less than or equal to the critical respiratory wave harmonic ratio, and adjust the gradual duration of the audio change process based on the number of triggers.

[0025] The analysis module is also used to extend the gradation duration based on the comparison result between the number of triggers and the preset number of triggers, and the extension of the gradation duration is positively correlated with the number of triggers;

[0026] The trigger count is the number of times the respiratory wave harmonic ratio is less than or equal to the critical respiratory wave harmonic ratio, and the gradation duration is the duration for the audio to change from one rhythm frequency to another.

[0027] Furthermore, the analysis module is also used to determine that the audio intervention process is unqualified based on the comparison result that the R-wave amplitude variation coefficient is greater than the second preset R-wave amplitude variation coefficient, and to adjust the gradation duration based on the amplitude variation coefficient offset value;

[0028] The amplitude variation coefficient offset value is the difference between the R-wave amplitude variation coefficient and the second preset R-wave amplitude variation coefficient.

[0029] Furthermore, the analysis module is also used to extend the gradation duration based on the comparison result between the amplitude variation coefficient offset value and the preset amplitude variation coefficient offset value, and the extension of the gradation duration is positively correlated with the amplitude variation coefficient offset value.

[0030] Compared with existing technologies, the beneficial effects of the smart sleep-aid headphones based on electrocardiogram (ECG) monitoring of the present invention are as follows: The original ECG signal is acquired and analyzed to obtain the RR interval sequence and R wave amplitude sequence. Based on the RR interval sequence, heart rate variability analysis is performed to obtain the RR interval standard deviation, the first signal frequency, and the second signal frequency, thereby determining the corresponding sleep stage. Based on the sleep stage, corresponding audio is played to intervene in the wearer's sleep process. The R wave amplitude sequence is analyzed to obtain the R wave amplitude variation coefficient, thereby determining whether the audio intervention process is qualified. Based on the determined results, an intervention strategy is selected: if qualified, the current audio playback is maintained; if unqualified, dynamic adjustments are made to correct the audio intervention process. This can specifically improve the consistency between the wearer's ECG frequency and the rhythm frequency of the audio played by the headphones, thereby achieving sleep aid.

[0031] Furthermore, this invention accurately determines each corresponding stage in the sleep process based on the standard deviation of the RR interval, the first signal frequency, and the second signal frequency, and selects the corresponding audio type according to the corresponding stage, thereby achieving precise sleep aid and improving the sleep aid efficiency of smart sleep aid headphones.

[0032] Furthermore, the present invention can analyze the audio intervention process based on the comparison result of the R-wave amplitude variation coefficient and the preset R-wave amplitude variation coefficient, thereby determining whether the current audio intervention process is qualified. In addition, when making the comparison judgment, the audio intervention process can be further judged based on the comparison result of the obtained respiratory wave harmonic noise ratio and the critical respiratory wave harmonic noise ratio, thereby improving the accuracy of the judgment process and accurately analyzing the audio intervention process of the smart sleep aid headphones.

[0033] Furthermore, when the present invention determines that there is an abnormality in the audio intervention process, it can accurately reduce the rhythm frequency of the audio based on the comparison result of the harmonic noise ratio difference and the preset harmonic noise ratio difference, thereby reducing the impact of the audio played by the smart sleep aid headphones on the wearer and stabilizing the wearer's sleep state.

[0034] Furthermore, when the audio intervention process is determined to be unqualified, the present invention can extend the transition duration based on the comparison result of the number of triggers and the preset number of triggers; or, it can extend the transition duration based on the amplitude variation coefficient offset value and the preset amplitude variation coefficient offset value, thereby adaptively and smoothly switching the rhythm frequency of the audio, thereby reducing the stimulation to the wearer during sleep, improving the consistency between the wearer's electrocardiogram frequency and the rhythm frequency of the audio played by the headphones, and helping to promote sleep. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of a module of an intelligent sleep aid headphone based on electrocardiogram monitoring according to the present invention;

[0036] Figure 2 This is a schematic diagram illustrating the workflow of a smart sleep aid headphone based on electrocardiogram monitoring according to the present invention.

[0037] Figure 3 This is a logic diagram for determining the corresponding sleep stage based on the ratio of the standard deviation of the RR interval to the frequency in this invention.

[0038] Figure 4 This is a logic diagram for determining whether the audio intervention process is qualified and the corresponding processing based on the coefficient of variation of R-wave amplitude in this invention. Detailed Implementation

[0039] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0040] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0041] It should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the term "connection" should be interpreted broadly. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0042] Please see Figure 1The diagram shown illustrates the modules of the smart sleep aid headphones based on electrocardiogram (ECG) monitoring in this embodiment. The smart sleep aid headphones in this embodiment include an ECG acquisition module, an R-wave detection module, a feature extraction module, a sleep stage module, an audio playback module, an analysis module, and a smart control module. The ECG acquisition module is used to acquire the wearer's raw ECG signal in real time and perform preprocessing to obtain a processed ECG signal; the R-wave detection module is connected to the ECG acquisition module and is used to perform R-wave detection analysis based on the processed ECG signal to obtain the RR interval sequence and R-wave amplitude sequence; the feature extraction module is connected to the R-wave detection module and is used to perform heart rate variability analysis based on the RR interval sequence to obtain the RR interval standard deviation, the first signal frequency, and the second signal frequency, and to perform analysis based on the R-wave amplitude sequence to obtain the R-wave amplitude variation coefficient; the sleep stage module is connected to the feature extraction module and is used to perform analysis based on the RR interval standard deviation, the first signal frequency, and the second signal frequency, and to obtain the R-wave amplitude variation coefficient based on the R-wave amplitude sequence; the sleep stage module is connected to the feature extraction module and is used to perform analysis based on the RR interval standard deviation, the first signal frequency, the second signal frequency, and the second signal frequency. The frequency and the second signal frequency determine the corresponding sleep stage; the audio playback module is connected to the sleep stage module to pre-store several different types of audio and determine the baseline value of the electrocardiogram frequency based on the sleep stage, so as to play the corresponding type of audio to perform audio intervention on the wearer; the analysis module is connected to the feature extraction module and the audio playback module respectively, and is used to determine whether the audio intervention process is qualified based on the changes in the R wave amplitude variation coefficient and the respiratory wave harmonic noise ratio; the intelligent control module is connected to the analysis module and the audio playback module respectively, and is used to maintain the playback of the current audio or dynamically adjust the rhythm and frequency of the audio based on the results determined by the audio intervention process.

[0043] Please see Figure 2 The diagram shown illustrates the workflow of the smart sleep aid headphones based on electrocardiogram monitoring in this embodiment. The workflow in this embodiment includes at least the following steps:

[0044] S1: Real-time acquisition of the wearer's raw electrocardiogram (ECG) signal and preprocessing to obtain the processed ECG signal;

[0045] S2: Based on the processed ECG signal, R-wave detection and analysis are performed to obtain the RR interval sequence and R-wave amplitude sequence;

[0046] S3: Heart rate variability analysis based on RR interval sequence to obtain RR interval standard deviation, first signal frequency and second signal frequency, and analysis based on R wave amplitude sequence to obtain R wave amplitude variation coefficient;

[0047] S4: Determine the corresponding sleep stage based on the standard deviation of the RR interval, the first signal frequency, and the second signal frequency;

[0048] S5: Determine the baseline value of ECG frequency based on the sleep stage, and perform audio intervention on the wearer by playing corresponding types of audio, where the corresponding audio consists of several pre-stored audio of different types;

[0049] S6: Determine whether the audio intervention process is qualified based on the changes in the R-wave amplitude variation coefficient and the respiratory wave harmonic-to-noise ratio;

[0050] S7: Based on the results determined during the audio intervention process, maintain the current audio playback or dynamically adjust the rhythm and frequency of the audio.

[0051] Specifically, in this embodiment, the ECG acquisition module can use a bioelectric sensor to acquire the wearer's raw ECG signal in real time, and then preprocess the acquired signal. The preprocessing process includes power frequency interference suppression, baseline drift removal, electromyographic noise suppression, motion artifact compensation, bandpass filtering, etc. The preprocessing process is existing technology and will not be described in detail here. The bioelectric sensor can be a concha-tragus electrode pair. The ECG acquisition module uses the configured concha-tragus electrode to acquire bioelectric signals from the ear, thereby realizing the real-time acquisition of the raw ECG signal. The R-wave detection module can employ a dual-processing engine (or a dual-core processor). The first processing engine provides channel 1 to process the RR interval stream to obtain the RR interval sequence and performs abnormal heartbeat correction. The second processing engine provides channel 2 to process the R-wave amplitude stream to obtain the R-wave amplitude sequence and performs baseline drift compensation and amplitude normalization. Then, based on the RR interval sequence, time-domain analysis of heart rate variability is performed to obtain the RR interval standard deviation, and frequency-domain analysis of heart rate variability is performed to obtain the first signal frequency and the second signal frequency. In this embodiment, the first signal frequency is sympathetic / vagal mixed activity, reflecting cardiovascular pressure regulation, and its range is 0.04–0.15 Hz. The second signal frequency is pure vagal activity, quantifying the respiratory-cardiac coupling strength, and its range is 0.16–0.4 Hz. The balance between the sympathetic and vagal nerves is quantified by the first signal frequency / second signal frequency. Analysis is also performed based on the R-wave amplitude sequence to obtain the R-wave amplitude variation coefficient, which quantifies the intensity of respiratory sinus arrhythmia, thereby reflecting the efficiency of vagal tone in regulating the cardiovascular system. The sleep stage module acquires the standard deviation of the RR interval, the first signal frequency (LF), and the second signal frequency (HF), and compares these three parameters to determine the wearer's current sleep stage. Based on this, it identifies the baseline ECG frequency for that stage, providing a basis for subsequent audio playback. The baseline ECG frequency includes the resting heart rate range and dynamic heart rate variability. The audio playback module utilizes an audio processor with a pre-stored audio library containing various frequency mixes, ambient sounds, and trigger sounds. It then plays corresponding audio based on the sleep stage to aid sleep. The analysis module acquires the R-wave amplitude variation coefficient in real time. After detecting the start of audio playback, it uses the newly acquired R-wave amplitude variation coefficient from the feature extraction module, combined with changes in the respiratory wave harmonic-to-noise ratio, to assess the effectiveness of the audio intervention. Based on the results determined by the analysis module, the intelligent control module selects the corresponding intervention strategy. If the results are satisfactory, it maintains the current audio playback; if the results are unsatisfactory, it makes dynamic adjustments to correct the audio intervention process, thereby achieving intelligent sleep aid and helping the wearer improve sleep quality.

[0052] Please see Figure 3 As shown, this is the logic diagram for determining the corresponding sleep stage based on the standard deviation of the RR interval and the frequency ratio in this embodiment. The sleep stage module is also used to determine the corresponding sleep stage based on the comparison results of the standard deviation of the RR interval and the preset standard deviation of the RR interval, and the comparison results of the frequency ratio and the preset frequency ratio; wherein, the sleep stage includes the sleep onset stage, the light sleep stage, and the deep sleep stage, and the frequency ratio is the ratio of the first signal frequency to the second signal frequency.

[0053] In this embodiment, the preset frequency ratio G0 is determined based on the physiological statistics of the user's historical sleep data. Specifically, under conditions of natural sleep and a stable sleep environment, ECG monitoring data from multiple sleep cycles are collected, and stable deep sleep stages are simultaneously identified using a gold standard (such as polysomnography, PSG). The LF / HF ratios (i.e., the ratio of the first signal frequency to the second signal frequency) corresponding to these identified deep sleep stages are extracted to form a deep sleep sample set. By calculating the upper limit or a specific quantile (such as the 95th percentile) of this sample set, the obtained statistical value is established as the preset frequency ratio G0, which is used as a threshold reference for deep sleep characteristics in subsequent real-time sleep stage determination. The preset RR interval standard deviation M0 is determined based on the physiological statistics of the user's historical sleep data. Specifically, under conditions of natural sleep and a stable sleep environment, ECG monitoring data from multiple sleep cycles are collected, and stable deep sleep stages are simultaneously identified using a gold standard (such as polysomnography, PSG). The RR interval standard deviations corresponding to these identified deep sleep stages are extracted to form a deep sleep sample set. By calculating the statistical characteristic values ​​(such as the mean or median) of the sample set, the obtained statistical value is established as the preset RR interval standard deviation M0, which serves as the key dynamic threshold for distinguishing between light sleep and deep sleep in subsequent real-time sleep stage determination.

[0054] Specifically, to accurately distinguish the current sleep stage of the wearer, the preset RR interval standard deviation M0 can be divided into a first preset RR interval standard deviation M1 and a second preset RR interval standard deviation M2, and the preset frequency ratio G0 can be divided into a first preset frequency ratio G1 and a second preset frequency ratio G2. For example, M1 = 25ms, M2 = 60ms, G1 = 1.0, and G2 = 1.5 are set. The comparison process based on the RR interval standard deviation M with M1 and M2, and based on the frequency ratio G with G1 and G2, is as follows:

[0055] If M is greater than M1 and less than or equal to M2, and G is greater than G1 and less than or equal to G2, the wearer's current sleep stage can be determined to be light sleep. If M is greater than M2 and G is greater than G2, the wearer's current sleep stage can be determined to be deep sleep. If neither M nor G falls into the first two categories, the wearer's current sleep stage is determined to be the sleep onset stage. At this point, the sleep stage label is output to provide a reference for subsequent playback of the corresponding audio type.

[0056] Furthermore, the audio playback module is also used to play audio of the type corresponding to the current sleep stage. Different types of audio have different rhythm frequencies, and each corresponds to a sleep stage.

[0057] Specifically, in this embodiment, when it is determined that the wearer is in the sleep stage, the wearer is guided to fall asleep by playing corresponding audio; when it is determined that the wearer is in the light sleep stage, the wearer is encouraged to fall asleep by playing corresponding audio; and when it is determined that the wearer is in the deep sleep stage, the wearer is encouraged to fall asleep by playing corresponding audio.

[0058] Please see Figure 4 As shown, this is the logic diagram for determining whether the audio intervention process is qualified and the corresponding processing based on the R-wave amplitude variation coefficient in this embodiment. The analysis module is also used to determine whether the audio interference process is qualified based on the comparison results of the R-wave amplitude variation coefficient and the preset R-wave amplitude variation coefficient, and the comparison results of the respiratory wave harmonic noise ratio and the critical respiratory wave harmonic noise ratio; wherein, if the R-wave amplitude variation coefficient is greater than the first preset R-wave amplitude variation coefficient and less than or equal to the second preset R-wave amplitude variation coefficient, the audio intervention process is re-determined based on the comparison results of the respiratory wave harmonic noise ratio and the critical respiratory wave harmonic noise ratio; wherein, the preset R-wave amplitude variation coefficient includes the first preset R-wave amplitude variation coefficient and the second preset R-wave amplitude variation coefficient.

[0059] Specifically, the feature extraction module can also obtain the respiratory wave harmonic-to-noise ratio J when analyzing the R-wave amplitude sequence. A baseline value A3 for the R-wave amplitude variation coefficient is set, where A3 represents the mean of the R-wave amplitude variation coefficient A obtained in the wearer's resting state within 5 minutes before falling asleep. The preset R-wave amplitude variation coefficient A0 can be divided into a first preset R-wave amplitude variation coefficient A1 and a second preset R-wave amplitude variation coefficient A2.

[0060] In this embodiment, the preset R-wave amplitude variation coefficient A0 is determined based on the user's individual physiological baseline statistics in a resting state. Specifically, before the audio intervention begins, the user is in a relaxed and awake resting state, and their electrocardiogram (ECG) signal is collected over a continuous period (e.g., 5 minutes). This signal is analyzed by a feature extraction module to obtain a series of R-wave amplitude variation coefficients within that period. The average value of this series is calculated and established as the baseline value A3. Subsequently, based on the baseline value A3, calculations are performed according to a preset ratio, and the calculated values ​​are established as the first preset R-wave amplitude variation coefficient A1 and the second preset R-wave amplitude variation coefficient A2, respectively. Thus, A1 and A2, dynamically determined based on the individual baseline A3, together constitute the personalized threshold interval A0 used to determine whether the audio intervention is qualified.

[0061] For example, let's set A1 = 0.85 × A3 and A2 = 1.15 × A3. The comparison process between A and A1 and A2 is as follows:

[0062] If A is less than or equal to A1, it indicates that the current vagus nerve-mediated respiratory-electrocardiographic coupling intensity has not reached the physiological activation threshold, meaning that the currently played audio does not interfere with the wearer's sleep state. Therefore, the current audio intervention process can be judged as qualified. If A is greater than A1 and less than or equal to A2, it indicates that the wearer is physiologically in an excessive or interfered state. At this time, the wearer may be in a state of respiratory distortion or semi-inhibition of the nervous system. In this case, it is impossible to accurately determine whether the current audio intervention process is qualified based on the standard deviation of the RR interval M, the first signal frequency, and the second signal frequency. Therefore, by introducing a new parameter, the respiratory wave harmonic noise ratio J, the current stage can be re-evaluated based on the changes in the respiratory wave harmonic noise ratio J to accurately determine whether there is any abnormality in the current audio intervention process. The new parameter can avoid relying on a single analysis and analyze from a multi-dimensional perspective to improve the accuracy of the judgment process.

[0063] If A is greater than A2, it indicates that the wearer's physiological state is getting closer to the activation threshold, and the wearer is approaching a state of wakefulness. Therefore, the current audio intervention process can be determined to be unqualified. At this time, the difference between the R-wave amplitude variation coefficient A and the second preset R-wave amplitude variation coefficient A2 can be obtained and recorded as the amplitude variation coefficient offset value N. The gradual duration is adjusted through N. The R-wave amplitude variation coefficient A at this time is the new R-wave amplitude variation coefficient obtained by the feature extraction module after analyzing the R-wave amplitude sequence after the audio intervention. By quantifying the regulatory efficiency of vagal tone on the cardiovascular system, the shortcomings of existing analyses based solely on heart rate variability, such as response lag and susceptibility to interference, are addressed, providing a fast and reliable basis for determining the audio intervention process.

[0064] Furthermore, the analysis module is also used to adjust the rhythm frequency of the currently played audio based on the comparison result between the respiratory wave harmonic noise ratio and the critical respiratory wave harmonic noise ratio; wherein, if the respiratory wave harmonic noise ratio is less than or equal to the critical respiratory wave harmonic noise ratio, it is determined that there is an abnormality in the audio intervention process, and the rhythm frequency of the audio is adjusted based on the harmonic noise ratio difference; wherein, the harmonic noise ratio difference is the difference between the critical respiratory wave harmonic noise ratio and the respiratory wave harmonic noise ratio.

[0065] Specifically, by setting a critical respiratory wave harmonic ratio J0 corresponding to the respiratory wave harmonic ratio J, the current audio intervention process can be re-evaluated based on the comparison between J and J0.

[0066] In this embodiment, the critical respiratory wave harmonic ratio J0 is determined based on the physiological statistical correlation of historical effective sleep aid data. Specifically, under the condition that audio intervention is effective, data from multiple monitoring periods during historical sleep aid processes are collected. Periods with stable physiological states (e.g., stable ECG signals and sleep stages determined to be stable light or deep sleep) are selected, and the corresponding respiratory wave harmonic ratio values ​​for these periods are extracted to form an effective sample set. By calculating the lower limit or a specific quantile (e.g., the 10th percentile) of this sample set, the obtained statistical value is established as the critical respiratory wave harmonic ratio J0.

[0067] For example, J0 is set to 12dB. The comparison process between J and J0 is as follows: if J is less than or equal to J0, it indicates that the wearer's current breathing is relatively unstable, and the audio intervention process is abnormal at this time, requiring dynamic adjustment. The rhythm frequency of the audio can be adjusted based on the harmonic-to-noise ratio difference L to reduce the impact of the audio intervention on the wearer's sleep state; if J is greater than J0, it indicates that the wearer's current breathing is relatively stable, and the audio intervention process is qualified at this time, and the current audio playback can be maintained.

[0068] Furthermore, the analysis module is also used to determine the reduced rhythm frequency of the audio based on the comparison result between the harmonic noise ratio difference and the preset harmonic noise ratio difference. The reduction in the rhythm frequency of the audio is positively correlated with the harmonic noise ratio difference.

[0069] Specifically, the harmonic-to-noise ratio difference L is the difference between the critical respiratory wave harmonic-to-noise ratio J0 and the respiratory wave harmonic-to-noise ratio J. L decreases as J increases. A larger J indicates more unstable breathing in the wearer, requiring a smaller reduction in the corresponding audio frequency to avoid excessive reduction affecting the wearer. In other words, a smaller L corresponds to a smaller reduction in the corresponding frequency. Therefore, the reduction in audio frequency is positively correlated with L. By reducing the frequency of the corresponding audio, vagal nerve tone is enhanced, thereby stabilizing breathing and improving the wearer's sleep quality. A preset harmonic-to-noise ratio difference L0 is set, and the adjustment range of audio frequency reduction is determined based on the comparison between L and L0.

[0070] In this embodiment, the preset harmonic-to-noise ratio (HNR) difference L0 is determined based on the physiological statistical correlation of historical intervention data. Specifically, under the condition that the audio intervention effectively regulates the physiological state, data from multiple monitoring periods during historical sleep aid processes are collected. Periods in which the wearer's respiratory wave HNR stabilizes after reducing the audio rhythm frequency are selected. The HNR difference corresponding to these periods before the intervention (i.e., the difference between the critical respiratory wave HNR and the actual respiratory wave HNR) is extracted to form an effective regulation sample set. By calculating the upper limit or a specific quantile (such as the 75th percentile) of this sample set, the obtained statistical value is determined as the preset HNR difference L0.

[0071] To more accurately determine the reduction adjustment range of the audio frequency, the preset harmonic-to-noise ratio difference L0 can be divided into a first preset harmonic-to-noise ratio difference L1 and a second preset harmonic-to-noise ratio difference L2. For example, L1 = 2dB and L2 = 4dB. The comparison process between L and L1 and L2 is as follows:

[0072] If L is less than or equal to L1, the analysis module generates a first audio frequency adjustment command, and the intelligent control module reduces the original audio frequency by 5% based on this command. If L is greater than L1 and less than or equal to L2, the analysis module generates a second audio frequency adjustment command, and the intelligent control module reduces the original audio frequency by 10% based on this command. If L is greater than L2, the analysis module generates a third audio frequency adjustment command, and the intelligent control module reduces the original audio frequency by 15% based on this command. It is understandable that the reduction in audio frequency can also be set to other suitable values; for example, when L is less than or equal to L1, the original audio frequency can be reduced by 6%.

[0073] Furthermore, the analysis module is also used to determine, within a preset duration after reducing the rhythm frequency of the audio, the wearer's current sleep process is in an abnormal and disordered state based on the comparison result that the harmonic noise ratio difference is greater than the second preset harmonic noise ratio difference; the intelligent control module is also used to emit interference audio to interrupt the wearer's sleep state, wherein the preset harmonic noise ratio difference includes the second preset harmonic noise ratio difference.

[0074] Specifically, in this embodiment, after the corresponding adjustment of the audio frequency is completed, the harmonic noise ratio difference L is re-acquired within a preset time period and compared with the second preset harmonic noise ratio difference L2. If L is still greater than L2, it indicates that after the adjustment of the audio frequency, the respiratory wave harmonic noise ratio J is still much smaller than the critical respiratory wave harmonic noise ratio J0. Thus, it can be determined that the wearer is in a state of abnormal and disordered breathing during the current sleep process. At this time, in order to reduce the safety risks to the wearer, the intelligent control module can issue interference to interrupt the wearer's sleep state. The interference mode includes issuing a first-level intervention (reminder) and a second-level intervention (wake-up) through the ECG acquisition module, and only when the medical mode is configured and the user authorizes it in advance can a very small amount of transcutaneous electrical nerve stimulation (TENS) be enabled, and the intensity must be strictly medically verified, thereby interrupting the wearer's sleep. The interference mode can be implemented using existing technology, which will not be elaborated here.

[0075] Furthermore, the analysis module is also used to determine that the audio intervention process is unqualified within a preset duration after reducing the rhythm frequency of the audio, based on the comparison results of the respiratory wave harmonic ratio being less than or equal to the critical respiratory wave harmonic ratio, and to adjust the gradation duration in the audio change process based on the number of triggers; the analysis module is also used to extend the gradation duration based on the comparison results of the number of triggers and the preset number of triggers, and the extension of the gradation duration is positively correlated with the number of triggers; where the number of triggers is the number of times the respiratory wave harmonic ratio is less than or equal to the critical respiratory wave harmonic ratio, and the gradation duration is the duration taken for the audio to change from one rhythm frequency to another.

[0076] Specifically, in this embodiment, the respiratory wave harmonic-to-noise ratio J is acquired several times during the wearer's sleep stage. After adjusting the rhythm frequency of the corresponding audio by comparing the harmonic-to-noise ratio difference L with the preset harmonic-to-noise ratio difference L0, the analysis module acquires a new respiratory wave harmonic-to-noise ratio J again and compares it with J0 to analyze the audio intervention process. If there is still a situation where J is less than or equal to J0, it can be determined that the current audio intervention process is unqualified. The number of triggers Y can be acquired and the gradation duration can be extended based on the number of triggers Y. By extending the gradation duration, the process of the audio transitioning from one rhythm frequency to another becomes smoother, thereby reducing the stimulation received by the wearer during sleep and improving the wearer's sleep quality.

[0077] In this embodiment, a preset trigger count Y0 corresponding to the trigger count Y can be set, and the gradation duration can be adjusted by comparing the results of Y and Y0.

[0078] The preset trigger count Y0 is determined based on the statistical correlation of historical intervention and adjustment data. Specifically, during the dynamic adjustment of the audio rhythm frequency, multiple adjustment cycles from historical sleep aid processes are collected. Cycles in which the wearer's physiological state (e.g., respiratory wave harmonic-to-noise ratio) tends to stabilize after adjustment are selected. The number of consecutive abnormal triggers corresponding to these cycles before reaching stability is extracted, forming an effective adjustment sample set. The statistical distribution and specific quantiles (e.g., the 50th and 80th percentiles) of this sample set are calculated, and the resulting statistical value is determined as the preset trigger count Y0.

[0079] To more accurately determine the adjustment range of the gradient duration, Y0 can be divided into a first preset trigger count Y1 and a second preset trigger count Y2, with Y1=2 times and Y2=4 times. The comparison process between Y and Y1 and Y2 is as follows:

[0080] If Y is less than or equal to Y1, the analysis module generates a first gradual duration adjustment command, and the intelligent control module extends the original gradual duration by 10% based on this command. If Y is greater than Y1 and less than or equal to Y2, the analysis module generates a second gradual duration adjustment command, and the intelligent control module extends the original gradual duration by 15% based on this command. If Y is greater than Y2, the analysis module generates a third gradual duration adjustment command, and the intelligent control module extends the original gradual duration by 20% based on this command. It is understood that the extension range of the gradual duration can also be set to other suitable values; for example, when Y is greater than Y2, the original gradual duration can be extended by 25%. It should be noted that adjusting the gradual duration will not negatively interfere with the wearer's sleep process. The maximum gradual duration is set to 10 seconds.

[0081] Furthermore, the analysis module is also used to determine that the audio intervention process is unqualified based on the comparison result that the R-wave amplitude variation coefficient is greater than the second preset R-wave amplitude variation coefficient, and to adjust the gradation duration based on the amplitude variation coefficient offset value; wherein, the amplitude variation coefficient offset value is the difference between the R-wave amplitude variation coefficient and the second preset R-wave amplitude variation coefficient.

[0082] Furthermore, the analysis module is also used to extend the gradation duration based on the comparison result between the amplitude variation coefficient offset value and the preset amplitude variation coefficient offset value. The extension of the gradation duration is positively correlated with the amplitude variation coefficient offset value.

[0083] Specifically, in this embodiment, when the R-wave amplitude variation coefficient A is greater than the second preset R-wave amplitude variation coefficient A2, the current audio intervention process can be determined to be unqualified. By obtaining the amplitude variation coefficient offset value N and setting a corresponding preset amplitude variation coefficient offset value N0, the fading duration is adjusted based on the comparison result between N and N0. To more accurately determine the adjustment range of the fading duration, N0 can be divided into a first preset amplitude variation coefficient offset value N1 and a second preset amplitude variation coefficient offset value N2, with N1 = 0.05 × A3 and N2 = 0.15 × A3. The comparison process between N and N1 and N2 is as follows:

[0084] If N is less than or equal to N1, the analysis module generates a fourth gradation duration adjustment command, and the intelligent control module extends the original gradation duration by 25% based on this command. If N is greater than N1 and less than or equal to N2, the analysis module generates a fifth gradation duration adjustment command, and the intelligent control module extends the original gradation duration by 30% based on this command. If N is greater than N2, the analysis module generates a sixth gradation duration adjustment command, and the intelligent control module extends the original gradation duration by 35% based on this command. Similarly, the extension range of the gradation duration can also be set to other values ​​that meet the requirements. For example, when N is greater than N2, the original gradation duration can be extended by 40%.

[0085] All technologies not mentioned in the above embodiments are existing technologies. It is understood that no specific limitation is made to any preset parameter or critical parameter in the embodiments of the present invention, and the above values ​​are not limited thereto. Those skilled in the art can adjust the preset parameters or critical parameters accordingly based on actual needs, analysis of historical data, or equipment usage.

[0086] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0087] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A smart sleep aid headphone based on electrocardiogram monitoring, characterized in that, include: The ECG acquisition module is used to acquire the wearer's raw ECG signal in real time and preprocess it to obtain the processed ECG signal; The R-wave detection module is connected to the ECG acquisition module and is used to perform R-wave detection and analysis based on the processed ECG signal to obtain the RR interval sequence and the R-wave amplitude sequence. The feature extraction module, which is connected to the R-wave detection module, is used to perform heart rate variability analysis based on the RR interval sequence to obtain the RR interval standard deviation, the first signal frequency and the second signal frequency, and to perform analysis based on the R-wave amplitude sequence to obtain the R-wave amplitude variation coefficient. A sleep stage module, which is connected to the feature extraction module, is used to determine the corresponding sleep stage based on the standard deviation of the RR interval, the first signal frequency, and the second signal frequency. The sleep stage module is also used to determine the corresponding sleep stage based on the comparison results of the standard deviation of the RR interval and the preset standard deviation of the RR interval, and the comparison results of the frequency ratio and the preset frequency ratio. The sleep stages include the sleep onset stage, the light sleep stage, and the deep sleep stage, and the frequency ratio is the ratio of the first signal frequency to the second signal frequency. An audio playback module, which is connected to the sleep stage, is used to pre-store several different types of audio and determine the basic value of the electrocardiogram frequency based on the sleep stage, so as to play the corresponding type of audio to perform audio intervention on the wearer. An analysis module, which is connected to the feature extraction module and the audio playback module respectively, is used to determine whether the audio intervention process is qualified based on the changes in the R-wave amplitude variation coefficient and the respiratory wave harmonic-to-noise ratio. The analysis module is also used to determine whether the audio interference process is qualified based on the comparison results of the R-wave amplitude variation coefficient with the preset R-wave amplitude variation coefficient and the comparison results of the respiratory wave harmonic noise ratio with the critical respiratory wave harmonic noise ratio. If the R-wave amplitude variation coefficient is greater than the first preset R-wave amplitude variation coefficient and less than or equal to the second preset R-wave amplitude variation coefficient, the audio intervention process is re-determined based on the comparison result of the respiratory wave harmonic noise ratio and the critical respiratory wave harmonic noise ratio. A multi-parameter determination mechanism is constructed based on the R-wave amplitude variation coefficient and the respiratory wave harmonic-to-noise ratio. An intelligent control module, which is connected to the analysis module and the audio playback module respectively, is used to maintain the current playback of the audio or dynamically adjust the rhythm and frequency of the audio based on the results determined by the audio intervention process. Based on the re-acquisition of the harmonic-to-noise ratio difference L within a preset time period and comparison with the second preset harmonic-to-noise ratio difference L2, if L is greater than L2, it is determined that the wearer is in a state of abnormal and disordered breathing during the current sleep process, and interference is emitted through the intelligent control module to interrupt the wearer's sleep state. The interference modes include primary intervention, secondary intervention and transcutaneous electrical nerve stimulation emitted through the ECG acquisition module. The harmonic noise ratio difference is the difference between the critical respiratory wave harmonic noise ratio and the respiratory wave harmonic noise ratio, wherein the critical respiratory wave harmonic noise ratio is determined based on the physiological statistical correlation of historical effective sleep aid data.

2. The intelligent sleep aid headphones based on electrocardiogram monitoring according to claim 1, characterized in that, The audio playback module is also used to play audio of the type corresponding to the current sleep stage, wherein different types of audio have different rhythm frequencies and correspond one-to-one with the sleep stages.

3. The intelligent sleep aid headphones based on electrocardiogram monitoring according to claim 1, characterized in that, The analysis module is also used to adjust the rhythm frequency of the currently playing audio based on the comparison result between the respiratory wave harmonic ratio and the critical respiratory wave harmonic ratio. If the respiratory wave harmonic-to-noise ratio is less than or equal to the critical respiratory wave harmonic-to-noise ratio, it is determined that there is an abnormality in the audio intervention process, and the rhythm frequency of the audio is adjusted based on the harmonic-to-noise ratio difference.

4. The intelligent sleep aid headphones based on electrocardiogram monitoring according to claim 3, characterized in that, The analysis module is also used to determine the reduction of the rhythm frequency of the audio based on the comparison result between the harmonic-to-noise ratio difference and the preset harmonic-to-noise ratio difference. The reduction of the audio rhythm frequency is positively correlated with the harmonic-to-noise ratio difference.

5. The intelligent sleep aid headphones based on electrocardiogram monitoring according to claim 4, characterized in that, The analysis module is also used to determine, within a preset duration after reducing the rhythm frequency of the audio, the wearer's current sleep process is in an abnormal and disordered state based on the comparison result that the harmonic noise ratio difference is greater than the second preset harmonic noise ratio difference. The intelligent control module is also used to emit interfering audio to interrupt the wearer's sleep state, wherein the preset harmonic noise ratio difference includes a second preset harmonic noise ratio difference.

6. The intelligent sleep aid headphones based on electrocardiogram monitoring according to claim 5, characterized in that, The analysis module is also used to determine that the audio intervention process is unqualified within a preset duration after the rhythm frequency of the audio is reduced, based on the comparison result that the respiratory wave harmonic ratio is less than or equal to the critical respiratory wave harmonic ratio, and to adjust the gradual duration of the audio change process based on the number of triggers. The analysis module is also used to extend the gradation duration based on the comparison result between the number of triggers and the preset number of triggers, and the extension of the gradation duration is positively correlated with the number of triggers; The trigger count is the number of times the respiratory wave harmonic ratio is less than or equal to the critical respiratory wave harmonic ratio, and the gradation duration is the duration for the audio to change from one rhythm frequency to another.

7. The intelligent sleep aid headphones based on electrocardiogram monitoring according to claim 6, characterized in that, The analysis module is also used to determine that the audio intervention process is unqualified based on the comparison result that the R-wave amplitude variation coefficient is greater than the second preset R-wave amplitude variation coefficient, and to adjust the gradation duration based on the amplitude variation coefficient offset value. The amplitude variation coefficient offset value is the difference between the R-wave amplitude variation coefficient and the second preset R-wave amplitude variation coefficient.

8. The intelligent sleep aid headphones based on electrocardiogram monitoring according to claim 7, characterized in that, The analysis module is also used to extend the gradation duration based on the comparison result between the amplitude variation coefficient offset value and the preset amplitude variation coefficient offset value, and the extension of the gradation duration is positively correlated with the amplitude variation coefficient offset value.

Images