Method and device for regulating sleep-wake circadian rhythm disorder based on ultrasonic stimulation

Abstract

This application relates to the technical field of data analysis and processing, and discloses a method and device for regulating sleep-wake circadian rhythm disorders based on ultrasonic stimulation. The method includes: collecting users' sleep physiological data to construct a feature matrix; analyzing the sleep stage to obtain the disorder type; generating an ultrasonic stimulation parameter scheme; determining the brain region stimulation coordinates to generate spatial positioning data; predicting the user's feedback through a convolutional neural network; comparing the prediction with the actual feedback to adjust the scheme, and obtaining an optimized target stimulation scheme. This application realizes the accurate identification of sleep-wake circadian rhythm disorders, the generation of personalized stimulation parameters, the precise brain region positioning, the predictive evaluation, and the dynamic scheme optimization, thereby improving the regulation effect of ultrasonic stimulation on various sleep-wake circadian rhythm disorders.

Description

Technical Field

[0001] This application relates to the technical field of data analysis and processing, and particularly to a method and device for regulating sleep-wake circadian rhythm disorder based on ultrasonic stimulation. Background Art

[0002] Circadian rhythm sleep-wake disorder (CRSWD) is a common health problem, mainly manifested as the sleep-wake cycle being out of sync with the normal circadian rhythm, including various types such as phase delay syndrome, phase advance syndrome, non-24-hour sleep-wake syndrome, and irregular sleep-wake rhythm. The common symptoms of circadian rhythm sleep-wake disorder are difficulty falling asleep, difficulty maintaining sleep, and excessive daytime sleepiness. In severe cases, it can further affect health, damage social functions, work, life, study, and safety, etc. Currently, clinically, methods such as drug therapy, light therapy, melatonin supplementation, and cognitive behavioral therapy are mainly used for intervention. Although drug therapy takes effect quickly, it is prone to dependence and has various side effects; light therapy requires strict control of time and intensity and is inconvenient to use; the efficacy of melatonin supplementation varies from person to person, and it is difficult to ensure regular intake; cognitive behavioral therapy requires long-term adherence, and the patient compliance is poor. In recent years, non-invasive brain stimulation techniques such as repetitive transcranial magnetic stimulation (r-TMS) have been applied to the treatment of sleep disorders and have shown good efficacy. However, the large device volume, complex operation, and high cost limit its wide application.

[0003] The existing transcranial ultrasound stimulation (TUS) technology, as an emerging non-invasive brain regulation method, has gradually been applied to the field of neuromodulation due to its advantages such as moderate penetration depth, high spatial accuracy, and portable equipment. However, there are still obvious deficiencies in the current TUS technology for the treatment of circadian rhythm sleep-wake disorder: First, the setting of stimulation parameters lacks scientific basis and is mostly determined by experience, resulting in large differences in individualized treatment effects; second, there is a lack of an accurate target brain area localization mechanism, making it difficult to achieve precise stimulation of specific sleep regulation brain areas; third, there is a lack of a closed-loop feedback system and it is unable to dynamically adjust the treatment plan according to the actual response of the patient; fourth, the treatment effect prediction mechanism is not perfect, and it is difficult to evaluate the possible treatment effect before treatment. These problems seriously restrict the application effect and popularization value of ultrasonic stimulation technology in the regulation of circadian rhythm sleep-wake disorder. Summary of the Invention

[0004] This application provides a method and device for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation, which are used to achieve accurate identification of sleep-wake circadian rhythm disorders, generation of personalized stimulation parameters, precise brain region localization, predictive evaluation, and dynamic scheme optimization, so as to improve the regulation effect of ultrasound stimulation on various sleep-wake circadian rhythm disorders and overcome the technical defects such as arbitrary parameter setting, inaccurate localization, and lack of feedback mechanism in the prior art.

[0005] In the first aspect, this application provides a method for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation. The method for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation includes: collecting the respiratory frequency data, heart rate data, blood oxygen concentration data, and electromyogram data of the user during sleep, obtaining the facial image of the user, recording the electroencephalogram data, and constructing a sleep cycle feature matrix; analyzing the sleep stage distribution of the user according to the sleep cycle feature matrix, measuring the sleep latency and the number of nocturnal awakenings, and generating a determination result of the type of sleep-wake circadian rhythm disorder; generating a combination of ultrasound stimulation parameters, including stimulation frequency, stimulation intensity, stimulation duration, stimulation waveform, and stimulation focus position, based on the determination result of the type of sleep-wake circadian rhythm disorder, to obtain an ultrasound stimulation scheme; determining the key brain region stimulation coordinates, including the pineal gland region coordinates, hypothalamus region coordinates, prefrontal cortex region coordinates, and preoptic area coordinates, based on the ultrasound stimulation scheme, and generating spatial positioning data; performing user feedback prediction based on a convolutional neural network according to the ultrasound stimulation scheme through the spatial positioning data to obtain predicted feedback data; obtaining the actual feedback data of the user stimulated by the ultrasound stimulation scheme, generating feedback difference data based on the predicted feedback data and the actual feedback data, and correcting the ultrasound stimulation scheme according to the feedback difference data to obtain a target stimulation scheme.

[0006] In the second aspect, this application provides a device for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation. The device for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation includes:

[0007] A collection module, which is used to collect the respiratory frequency data, heart rate data, blood oxygen concentration data, and electromyogram data of the user during sleep, obtain the facial image of the user, record the electroencephalogram data, and construct a sleep cycle feature matrix;

[0008] An analysis module, which is used to analyze the sleep stage distribution of the user according to the sleep cycle feature matrix, measure the sleep latency and the number of nocturnal awakenings, and generate a determination result of the type of sleep-wake circadian rhythm disorder;

[0009] A generation module, configured to generate an ultrasonic stimulation parameter combination according to the determination result of the sleep-wake circadian rhythm disorder type, including stimulation frequency, stimulation intensity, stimulation duration, stimulation waveform, and stimulation focusing position, so as to obtain an ultrasonic stimulation plan;

[0010] A stimulation module, configured to determine key brain region stimulation coordinates according to the ultrasonic stimulation plan, including pineal region coordinates, hypothalamus region coordinates, prefrontal region coordinates, and preoptic area coordinates, and generate spatial positioning data;

[0011] A prediction module, configured to perform user feedback prediction based on a convolutional neural network according to the ultrasonic stimulation plan through the spatial positioning data, so as to obtain prediction feedback data;

[0012] A correction module, configured to obtain actual feedback data of a user stimulated by the ultrasonic stimulation plan, generate feedback difference data according to the prediction feedback data and the actual feedback data, and correct the ultrasonic stimulation plan according to the feedback difference data to obtain a target stimulation plan.

[0013] A third aspect of the present invention provides a computer device, including: a memory and at least one processor, wherein instructions are stored in the memory; the at least one processor calls the instructions in the memory to enable the computer device to execute the above-mentioned method for regulating sleep-wake circadian rhythm disorder based on ultrasonic stimulation.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, and when the instructions are run on a computer, the computer is enabled to execute the above-mentioned method for regulating sleep-wake circadian rhythm disorder based on ultrasonic stimulation.

[0015] In the technical solution provided by this application, the sleep cycle feature matrix constructed by collecting and integrating multi-dimensional physiological data greatly improves the accuracy of sleep disorder judgment. Compared with the analysis of a single data source, this matrix includes six heterogeneous data types: respiratory rate, heart rate, blood oxygen concentration, electromyogram data, facial images, and electroencephalogram data, providing a more comprehensive characterization of the sleep state and making the sleep stage recognition more accurate. Secondly, the method of segmentally setting ultrasonic stimulation parameters based on the determination result of the sleep-wake circadian rhythm disorder type selects a specific frequency range for different disorder types, effectively solving the problem of inaccurate setting of traditional empirical parameters. In particular, by targeted frequency selection (phase delay type: 200 - 500 kHz, phase advance type: 500 - 800 kHz, irregular type: 800 - 1200 kHz), the regulation accuracy of ultrasonic energy on the targeted brain area is significantly improved, enhancing the treatment pertinence. Moreover, the method for calculating the safe stimulation intensity considering the skull attenuation factor effectively balances the treatment effect and safety, ensuring that the ultrasonic energy can penetrate the skull to reach the target brain area without exceeding the tissue damage threshold, reducing the incidence of adverse reactions. In addition, the precise positioning technology of key brain areas improves the stimulation positioning accuracy to the millimeter level, which is a qualitative leap compared with the centimeter-level accuracy of traditional methods, directly enhancing the ultrasonic energy transfer efficiency by forming spatial positioning data. It is particularly worth emphasizing that the technical feature of applying a convolutional neural network for user feedback prediction in this solution makes an outstanding contribution. This model realizes the mapping from stimulation parameters to expected effects through a three-layer convolutional structure and a fully connected layer, reducing the treatment trial-and-error cost. The dynamic scheme adjustment mechanism based on feedback difference data forms a complete closed-loop control system. By comparing the predicted feedback with the actual feedback in real time, the parameter deviation is accurately calculated, and the ultrasonic stimulation scheme is optimized accordingly, overall achieving the personalization, precision, and efficiency of the sleep-wake circadian rhythm disorder regulation strategy analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0017] Figure 1 It is a schematic diagram of an embodiment of the method for regulating sleep-wake circadian rhythm disorder based on ultrasonic stimulation in an embodiment of this application;

[0018] Figure 2 It is a schematic diagram of an embodiment of the device for regulating sleep-wake circadian rhythm disorder based on ultrasonic stimulation in an embodiment of this application;

[0019] Figure 3 It is a schematic block diagram of the structure of a computer device in an embodiment of the present invention. Detailed implementation manners

[0020] The embodiments of the present application provide a method and a device for regulating sleep-wake circadian rhythm disorders based on ultrasonic stimulation. Terms such as "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present application are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments described here can be implemented in an order different from that shown or described here. In addition, the terms "comprising" or "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0021] For ease of understanding, the specific process of the embodiments of the present application will be described below. Please refer to Figure 1 , an embodiment of the method for regulating sleep-wake circadian rhythm disorders based on ultrasonic stimulation in the embodiments of the present application includes:

[0022] Step S101, collect the breathing frequency data, heartbeat data, blood oxygen concentration data and electromyogram data of the user during sleep, obtain the facial image of the user, record the electroencephalogram data, and construct a sleep cycle feature matrix;

[0023] Step S102, analyze the sleep stage distribution of the user according to the sleep cycle feature matrix, measure the sleep latency and the number of nocturnal awakenings, and generate a determination result of the sleep-wake circadian rhythm disorder type;

[0024] Step S103, generate an ultrasonic stimulation parameter combination according to the determination result of the sleep-wake circadian rhythm disorder type, including the stimulation frequency, stimulation intensity, stimulation duration, stimulation waveform and stimulation focusing position, and obtain an ultrasonic stimulation plan;

[0025] Step S104, determine the key brain region stimulation coordinates according to the ultrasonic stimulation plan, including the pineal gland region coordinates, hypothalamus region coordinates, prefrontal region coordinates and preoptic region coordinates, and generate spatial positioning data;

[0026] Step S105, perform user feedback prediction based on a convolutional neural network according to the ultrasonic stimulation plan through the spatial positioning data, and obtain predicted feedback data;

[0027] Step S106: Obtain the actual feedback data of the user stimulated by the ultrasonic stimulation scheme, generate feedback difference data based on the predicted feedback data and the actual feedback data, and modify the ultrasonic stimulation scheme according to the feedback difference data to obtain a target stimulation scheme.

[0028] It can be understood that the execution subject of this application can be a device for regulating sleep-wake circadian rhythm disorders based on ultrasonic stimulation, or a terminal or a server. Specifically, it is not limited here. In this embodiment of the application, the server is taken as the execution subject for illustration.

[0029] Specifically, when collecting multi-dimensional physiological data during the user's sleep, a multi-channel biosensing array is used to record the respiration rate, heartbeat, blood oxygen concentration, and electromyogram signals. These sensors are usually arranged under the user's mattress or in wearable devices without disturbing the user's normal sleep. At the same time, an infrared imaging device collects the user's facial images every hour, and a portable electroencephalogram device continuously records the electroencephalogram data. After noise elimination and smoothing processing of these raw data, time-domain features such as heart rate variability and frequency-domain features such as respiration rate power spectral density are extracted. The expression change features in the facial images and the energy distribution of four bands (delta wave, theta wave, alpha wave, and beta wave) in the electroencephalogram are integrated over time to form a sleep cycle feature matrix. This matrix contains the physiological state changes of the user throughout the sleep process. When analyzing the sleep stage distribution based on the sleep cycle feature matrix, the system extracts key time points from the electroencephalogram feature sequence and divides the sleep stage boundaries through the feature threshold segmentation method. For example, when the proportion of delta waves in the electroencephalogram exceeds 40% and the electromyogram activity decreases, it is determined as the deep sleep stage; when rapid eye movement appears and the electroencephalogram shows a similar waking state, it is determined as the REM sleep stage. The system calculates the time ratio of rapid eye movement sleep to non-rapid eye movement sleep, measures the time from waking to the first entry into light sleep (sleep latency), and identifies the number of nocturnal awakenings. These parameters are compared with the standard sleep pattern to generate a structural abnormality index, and a sleep disorder score is formed through weighted fusion. Finally, the system determines the type of the user's sleep-wake circadian rhythm disorder, such as phase delay syndrome, phase advance syndrome, or irregular sleep-wake rhythm.

[0030] When generating the ultrasonic stimulation parameter combination based on the determination result of the sleep-wake circadian rhythm disorder type, the system first extracts the parameter range of the corresponding type from the ultrasonic parameter base library. For the phase delay type disorder, the frequency range of 200 - 500 kHz is selected; for the phase advance type, 500 - 800 kHz; for the irregular type, 800 - 1200 kHz. The system calculates the safe stimulation intensity interval according to the frequency parameter, considering the skull attenuation factor to ensure that the ultrasonic energy can effectively penetrate the skull without causing tissue damage. The stimulation duration is set according to different sleep cycle regulation targets. For example, a longer time is selected for promoting deep sleep, and a shorter time is selected for regulating rapid eye movement sleep. The waveform type selection is also targeted: continuous wave is used for slow regulation, pulsed wave is used for rapid regulation, and modulated wave is used for mixed regulation. These parameters are integrated to form an ultrasonic stimulation plan.

[0031] When determining the stimulation coordinates of the key brain regions, the system first obtains the user's head tomographic scan image, extracts the skull structure features and brain tissue boundaries, and forms a three-dimensional brain structure model. In this model, the boundaries of the key brain regions such as the pineal gland region (responsible for melatonin secretion), the hypothalamus region (biological clock center), the prefrontal region (executive function regulation), and the preoptic area (sleep initiation) are marked, and the geometric center point of each region is calculated as the initial stimulation point. According to the stimulation focusing position requirements in the ultrasonic stimulation plan, the initial coordinates are precisely adjusted. The system also measures the thickness and density distribution of the user's skull at each stimulation point position, calculates the ultrasonic wave penetration path attenuation coefficient, and forms a skull sound transmission map. Finally, combining the precise stimulation point coordinates and the skull sound transmission map, the phase delay matrix and energy distribution function of the multi-element ultrasonic transducer are calculated to generate spatial positioning data.

[0032] When predicting the user feedback, the system combines the spatial positioning data with the ultrasonic stimulation plan to generate a stimulation target description matrix. Similar cases are extracted from the historical user database to construct a feedback feature vector library. The stimulation target description matrix is processed by a three-layer convolutional neural network. The first layer uses 32 3×3 convolutional kernels to extract low-level features (such as spatial position features), the second layer uses 64 3×3 convolutional kernels to extract intermediate features (such as parameter combination features), and the third layer uses 128 3×3 convolutional kernels to extract high-level features (such as stimulation effect features). After each layer of convolution, max pooling and batch normalization are performed to generate a multi-dimensional stimulation feature map. Through two fully connected layers (256 and 128 neurons), the high-dimensional features are mapped to the sleep parameter evaluation space to form a stimulation effect map. Finally, the correlation between the stimulation effect map and the feedback feature vector library is calculated to generate predicted feedback data.

[0033] The system obtains the actual feedback data of the user and modifies the solution. After the user receives the ultrasonic stimulation, the changes in sleep quality are monitored, including the frequency of sleep stage transitions, the proportion of deep sleep, and the duration of rapid eye movement sleep. At the same time, the changes in physiological data such as heart rate variability, respiratory rhythm, and body temperature curve are collected, as well as the subjective reports of the user on the difficulty of falling asleep, the feeling of nocturnal awakening, and the mental state in the morning. These data are integrated into the actual feedback data, compared with the predicted feedback data, and the feedback difference matrix is calculated. According to the deviation direction and amplitude of each parameter in the difference matrix, the ultrasonic stimulation solution is precisely adjusted to form the target stimulation solution.

[0034] For example, after a certain user is tested by this system, the sleep cycle characteristic matrix shows that the energy distribution of its δ wave is abnormal, the sleep latency reaches 45 minutes (normal value < 30 minutes), there are 6 nocturnal awakenings (normal value < 3 times), and the proportion of rapid eye movement sleep is only 15% (normal value 20 - 25%). The system determines it as a phase delay type sleep-wake circadian rhythm disorder and generates the initial ultrasonic stimulation parameters: frequency 350 kHz, intensity 0.5 W / cm², duration 5 minutes, continuous wave form, focused on the suprachiasmatic nucleus of the hypothalamus. It is predicted by the convolutional neural network that this solution will reduce the sleep latency to 32 minutes and the nocturnal awakenings to 4 times. After actual treatment, the user feedbacks that the sleep latency is 29 minutes and there are 3 nocturnal awakenings. The system calculates the feedback difference data and adjusts the ultrasonic parameters to: frequency 380 kHz, intensity 0.6 W / cm², duration 6 minutes, forming a more precise target stimulation solution and achieving an effective regulation of the sleep-wake circadian rhythm disorder.

[0035] In the embodiments of the present application, the sleep cycle feature matrix constructed by collecting and integrating multi-dimensional physiological data greatly improves the accuracy of sleep disorder judgment. Compared with single data source analysis, this matrix contains six heterogeneous data, namely respiratory rate, heart rate, blood oxygen concentration, electromyogram data, facial images, and electroencephalogram data, providing a more comprehensive representation of the sleep state and making the sleep stage identification more accurate. Secondly, the method of segmentally setting ultrasonic stimulation parameters based on the determination result of sleep-wake circadian rhythm disorder types selects specific frequency ranges for different disorder types, effectively solving the problem of inaccurate setting of traditional empirical parameters. In particular, by targeted frequency selection (phase delay type: 200 - 500 kHz, phase advance type: 500 - 800 kHz, irregular type: 800 - 1200 kHz), the regulation accuracy of ultrasonic energy on the targeted brain area is significantly improved, enhancing the treatment pertinence. Moreover, the method for calculating the safe stimulation intensity considering the skull attenuation factor effectively balances the treatment effect and safety, ensuring that the ultrasonic energy can penetrate the skull to reach the target brain area without exceeding the tissue damage threshold, reducing the incidence of adverse reactions. In addition, the precise positioning technology of key brain areas improves the stimulation positioning accuracy to the millimeter level, which is a qualitative leap compared with the centimeter-level accuracy of traditional methods, directly enhancing the ultrasonic energy transfer efficiency by forming spatial positioning data. It is particularly worth emphasizing that the technical feature of applying a convolutional neural network to predict user feedback in this solution makes an outstanding contribution. This model realizes the mapping from stimulation parameters to expected effects through a three-layer convolutional structure and a fully connected layer, reducing the treatment trial-and-error cost. The dynamic scheme adjustment mechanism based on feedback difference data forms a complete closed-loop control system. By comparing the predicted feedback with the actual feedback in real time, the parameter deviation is accurately calculated, and the ultrasonic stimulation scheme is optimized accordingly, overall realizing the personalization, precision, and efficiency of the analysis of sleep-wake circadian rhythm disorder regulation strategies.

[0036] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0037] (1) Collect respiratory rate data, heart rate data, blood oxygen concentration data, and electromyogram data of the user during seven consecutive days of sleep through a multi-channel biosensing array to obtain the original physiological data;

[0038] (2) Collect the facial images of the user once per hour during the user's sleep through an infrared imaging device to obtain a facial image sequence;

[0039] (3) Record the electroencephalogram data of the user during sleep through a portable electroencephalogram detection device to obtain electroencephalogram time series data;

[0040] (4) Perform noise elimination and data smoothing processing on the original physiological data to obtain the processed physiological data;

[0041] Extract time-domain features and frequency-domain features from the processed physiological data to obtain physiological feature values;

[0042] Extract facial expression change features from the facial image sequence to obtain facial expression feature values;

[0043] Separate the energy distributions of four bands, namely delta waves, theta waves, alpha waves, and beta waves, from the electroencephalogram time-series data to obtain electroencephalogram feature values;

[0044] Integrate the physiological feature values, the facial expression feature values, and the electroencephalogram feature values in time series to obtain a sleep cycle feature matrix.

[0045] Specifically, physiological data during a user's seven consecutive days of sleep is collected through a multi-channel biosensing array. This sensing array includes four dedicated sensors: a respiratory monitoring belt, an electrocardiogram monitoring electrode, a pulse oximeter, and an electromyogram electrode. These sensors are integrated into a portable device that the user can wear at home without having to go to a hospital or a sleep center. Respiratory rate data is collected through a chest and abdomen respiratory belt, recording the number of breaths per minute and the change in respiratory depth; heart rate data is collected through electrocardiogram electrodes attached to the chest, recording the heart rate and its variability; blood oxygen concentration data is collected through a finger clip pulse oximeter to monitor the change in oxygen saturation in the blood; electromyogram data is collected through surface electrodes attached to the jaw and legs, recording the muscle activity state. These data are collected once per second, forming a raw physiological data stream with high time resolution. At the same time, an infrared imaging device is installed at the head of the bed and triggered automatically once per hour during the user's sleep to capture the user's facial images. The infrared technology enables the device to work without interference in a completely dark environment and will not affect the user's normal sleep. 8-10 facial images are captured every night, and approximately 60 images are obtained in total within seven days, forming a facial image sequence. These images capture the facial expression changes of the user at different sleep stages, such as micro-expressions like rapid eye movement, frowning or relaxing of the eyebrows, etc.

[0046] The portable electroencephalogram detection device adopts dry electrode technology. Users only need to wear a device similar to a headband to record the electroencephalogram activities in the frontal, temporal, and parietal regions. The device collects data every 0.5 seconds, forming continuous electroencephalogram time-series data. These data contain the electroactivity information of the brain in different sleep stages, which is the key basis for judging the sleep depth and quality. The original physiological data often contains various interference signals, such as body movement artifacts, environmental electromagnetic interference, and device noise. The data is denoised by a digital filter. Specifically, band-pass filtering technology is used. For respiratory data, a filter of 0.1 - 0.5 Hz is applied, for electrocardiogram data, a filter of 0.5 - 40 Hz is applied, and for electromyogram data, a filter of 10 - 500 Hz is applied. The filtered data is then smoothed by the moving average method, taking the mean of 5 consecutive data points to replace the central point, thereby suppressing random fluctuations. These processes make the data clearer and more stable, facilitating subsequent feature extraction.

[0047] Time-domain features and frequency-domain features are extracted from the processed physiological data. Time-domain features include the average heart rate, standard deviation, maximum and minimum values, mean respiratory rate and coefficient of variation, average level and fluctuation amplitude of blood oxygen saturation, electromyogram activity intensity and burst times. Frequency-domain features are obtained through the fast Fourier transform (FFT), which converts the time-domain signal into the frequency domain and analyzes the energy distribution in different frequency bands. For example, in heart rate variability analysis, the powers of the very low-frequency component (VLF, <0.04 Hz), low-frequency component (LF, 0.04 - 0.15 Hz), and high-frequency component (HF, 0.15 - 0.4 Hz) are extracted, and the LF / HF ratio is calculated to reflect the sympathetic - parasympathetic nerve balance state. These eigenvalue values together constitute a multi-dimensional physiological feature vector.

[0048] The facial image sequence extracts the facial expression change features through computer vision technology. First, facial key point localization is performed, marking 68 feature points such as eyes, eyebrows, and mouth corners, and then the displacement amounts of these points between consecutive images are calculated to capture minute expression changes. The expression intensity is also quantified by defining a facial area grid and calculating the degree of grid deformation. In addition, eye movement is detected through brightness analysis to determine whether it is in the rapid eye movement period. These data form facial expression eigenvalue values, reflecting the subconscious state changes during sleep.

[0049] The electroencephalogram time-series data is decomposed into different frequency bands by wavelet transform: The δ wave (0.5 - 4 Hz) mainly appears in deep sleep, the θ wave (4 - 8 Hz) is common in light sleep, the α wave (8 - 12 Hz) mostly appears in the relaxed waking state, and the β wave (12 - 30 Hz) is related to mental activity. The proportion of each waveform in the total energy and the time variation law are calculated to form electroencephalogram eigenvalue values. These eigenvalue values are the core indicators for judging the sleep stage and depth.

[0050] Finally, synchronize and integrate the physiological feature values, facial expression feature values, and electroencephalogram feature values according to the timestamps to create a multi-dimensional data structure. Each time point has corresponding physiological states, facial expressions, and electroencephalogram activity data, forming a sleep cycle feature matrix that comprehensively describes the sleep process. This matrix is the data basis for subsequent analysis of the sleep stage distribution and determination of the type of sleep disorder.

[0051] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0052] (1) Extract the electroencephalogram feature sequence from the sleep cycle feature matrix, and divide the sleep stage boundaries by the feature threshold segmentation method to obtain a sleep stage time distribution map;

[0053] (2) Calculate the time ratio of rapid eye movement sleep to non-rapid eye movement sleep from the sleep stage time distribution map to obtain a sleep structure ratio value;

[0054] (3) Determine the time required to enter light sleep for the first time from the waking state according to the sleep stage time distribution map to obtain a sleep latency value;

[0055] (4) Identify the number of times of transitioning from deep sleep or light sleep to the waking state during the sleep process according to the sleep stage time distribution map to obtain the number of nocturnal awakenings;

[0056] (5) Compare the difference between the sleep structure ratio value and the standard sleep ratio range to obtain a structure abnormality index;

[0057] (6) Obtain a sleep disorder score by weighted fusion of the sleep latency value, the number of nocturnal awakenings, and the structure abnormality index;

[0058] (7) Generate a determination result of the sleep-wake circadian rhythm disorder type by comparing the sleep disorder score and the temporal pattern of the sleep stage time distribution map with the sleep disorder classification standard.

[0059] Specifically, an electroencephalogram (EEG) feature sequence is extracted from the sleep cycle feature matrix. This sequence contains data on the time-varying energy distributions of four frequency bands, namely delta waves, theta waves, alpha waves, and beta waves. The feature threshold segmentation method is a technique for classifying sleep stages based on setting specific thresholds for the energy ratios of these bands. In specific operations, when the proportion of delta waves exceeds 40%, it is determined as deep sleep (stage N3). When the proportion of theta waves exceeds 50% and the proportion of delta waves is less than 20%, it is determined as light sleep (stage N2). When both alpha waves and theta waves show significant activity and beta waves increase, it is determined as light sleep in stage N1. When the EEG shows a desynchronized pattern, the facial electromyogram shows a decrease in muscle tension, and rapid eye movements are detected by eye movement detection, it is determined as REM sleep. When beta waves are dominant and electromyogram activity increases, it is determined as the waking state. Judgments are made for the all-night sleep data period by period according to these criteria, and a sleep stage time distribution graph showing the changes in different sleep stages over time is plotted.

[0060] When calculating the time ratio of rapid eye movement (REM) sleep to non-rapid eye movement (NREM, including stages N1, N2, and N3) sleep from the sleep stage time distribution graph, first, the time periods determined as REM stages throughout the night are accumulated to obtain the total REM duration. Then, the time periods determined as stages N1, N2, and N3 are accumulated separately to obtain the total NREM duration. The two are divided to obtain the REM / NREM ratio. Normally, in normal sleep, this ratio should be between 0.2 and 0.25. Additionally, the proportion of deep sleep (N3) in the total sleep time is calculated, which should usually account for 15 - 25%. These values together constitute the sleep structure ratio value, reflecting the quality and structural integrity of sleep.

[0061] When determining the sleep latency value according to the sleep stage time distribution graph, the time interval from the start of the recording to the first time it is determined as any sleep stage (usually stage N1) is quantified. In specific operations, find the time point when the sleep recording starts, and then move forward along the time axis until the first time point determined as stage N1, and calculate the time difference between the two to obtain the sleep latency value. Normally, it should be less than 30 minutes. Exceeding this indicates difficulty falling asleep. When identifying the number of nocturnal awakenings, look for cases where any sleep stage (N1, N2, N3, or REM) changes to the waking state in the sleep stage time distribution graph, and count these transition points. Only awakenings lasting more than 30 seconds are counted as one awakening to exclude the influence of minor brief awakenings. The total number of such transitions throughout the night is counted to obtain the number of nocturnal awakenings, which is usually no more than 3 - 5 times in normal sleep.

[0062] When comparing the sleep structure proportion value with the standard sleep proportion range, a standardized difference scoring method is adopted. Taking the REM proportion as an example, if the measured value is 15%, the standard range is 20 - 25%, and the deviation is 5 - 10%, the standardized score is -1 to -2 points; if the deep sleep proportion is 8%, the standard range is 15 - 25%, and the deviation is 7 - 17%, the standardized score is -2 to -3 points. The standardized scores of each sleep structure index are added together to form a structure abnormality index. The higher this index, the further the sleep structure deviates from the normal standard.

[0063] When performing weighted fusion on the sleep latency value, the number of nocturnal awakenings, and the structure abnormality index, weights are assigned according to their different degrees of influence on sleep quality. Usually, the sleep structure abnormality has the greatest impact on health, with a weight of 0.5; frequent nocturnal awakenings come second, with a weight of 0.3; the sleep latency has a relatively smaller impact, with a weight of 0.2. After multiplying each of the three by the corresponding weight and adding them together, a comprehensive sleep disorder score is obtained. The score range is usually set as 0 - 100. The higher the score, the more severe the sleep disorder. According to the sleep disorder score and the temporal pattern of the sleep stage time distribution diagram, and in contrast to the sleep disorder classification standard, a determination result of the sleep - wake circadian rhythm disorder type is generated. In the specific determination, not only the magnitude of the disorder score is considered, but also the temporal pattern characteristics of the sleep stage distribution are analyzed. For example, if there is difficulty falling asleep with normal sleep in the later stage, a medium disorder score, and the sleep start time is more than 2 hours later than the normal time, it is determined as a phase delay syndrome; if both the sleep start and end times are too early, accompanied by difficulty maintaining sleep, it is determined as a phase advance syndrome; if the sleep - wake cycle length significantly deviates from 24 hours and is not fixed, it is determined as a non - 24 - hour sleep - wake syndrome; if the sleep is severely fragmented and the sleep stage transitions are frequent and irregular, it is determined as an irregular sleep - wake rhythm.

[0064] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0065] (1) Match the determination result of the sleep - wake circadian rhythm disorder type with the ultrasonic parameter base library, extract the parameter range table of the corresponding type, and obtain the initial parameter range;

[0066] (2) According to the initial parameter range, set the stimulation frequency in segments. Select the range of 200 - 500 kHz for the phase delay type disorder, the range of 500 - 800 kHz for the phase advance type disorder, and the range of 800 - 1200 kHz for the irregular type disorder to obtain the frequency parameter value;

[0067] (3) Based on the frequency parameter value, calculate the safe stimulation intensity interval, and at the same time consider the skull attenuation factor to obtain the stimulation intensity parameter value;

[0068] (4)For different sleep cycle regulation targets, set a gradient of stimulation duration, select a longer time parameter for promoting deep sleep, and select a shorter time parameter for regulating rapid eye movement sleep to obtain the stimulation duration parameter value;

[0069] (5)Based on the frequency parameter value and the stimulation intensity parameter value, select an appropriate waveform type, select a continuous wave for slow regulation, select a pulse wave for fast regulation, and select a modulated wave for mixed regulation to obtain the stimulation waveform parameter value;

[0070] (6)Integrate the frequency parameter value, the stimulation intensity parameter value, the stimulation duration parameter value, and the stimulation waveform parameter value to form an ultrasonic stimulation parameter combination, and set the stimulation focus position according to the rhythm regulation priority to obtain an ultrasonic stimulation scheme.

[0071] Specifically, match the determination result of the sleep-wake circadian rhythm disorder type with the ultrasonic parameter base library, which is a data set containing the parameter ranges corresponding to various sleep disorders, and associate the disorder type with the parameter range table through search and matching. In specific operations, when the determination result is phase delay syndrome, call the phase delay type parameter table; when it is phase advance syndrome, call the phase advance type parameter table; when it is irregular sleep-wake rhythm, call the irregular type parameter table. Each parameter table contains a frequency range, an intensity range, a duration range, waveform options, and a focus position priority list, which constitute the initial parameter range.

[0072] Based on the initial parameter range, the segmented setting of the stimulation frequency is based on the response characteristics of different types of disorders to ultrasonic frequencies. Select a range of 200 - 500 kHz for phase delay type disorders because lower frequency ultrasound can more effectively activate the suprachiasmatic nucleus of the hypothalamus, promote melatonin secretion, and help advance the sleep phase; select a range of 500 - 800 kHz for phase advance type disorders because medium frequency ultrasound can moderately inhibit pineal gland activity and delay the peak time of melatonin secretion; select a range of 800 - 1200 kHz for irregular type disorders because higher frequency ultrasound helps to stabilize the rhythmic activity of the sleep-wake center. Within the selected frequency range, further accurately select according to the severity of the patient's sleep disorder. The higher the disorder score, the closer the frequency setting is to the central value of the range, thereby obtaining the specific frequency parameter value.

[0073] When calculating the safe stimulation intensity interval based on the frequency parameter value, the skull attenuation factor needs to be considered, which involves a complex physical calculation process. The calculation formula for the safe stimulation intensity is:

[0074] ,

[0075] where, represents the safe stimulation intensity parameter value (W / cm²), Indicates the ultrasound bioeffect threshold intensity (W / cm²), represents the skull ultrasound attenuation coefficient ( ), which is frequency dependent, Indicates the thickness of the skull (cm), SF indicates the safety factor (usually 1.5-2). and frequency The relationship can be approximately expressed as:

[0076] ,

[0077] in, and are constants, 0.2 and 0.1 respectively, and p is the frequency exponent, usually 1.1. Obtained by head CT scan. For example, when the frequency parameter is 350kHz and the skull thickness is 0.6cm, the calculation is Approximately ,like 3W / cm², SF is 1.5, then the safe stimulation intensity Approximately 1.03W / cm². This ensures that the stimulation intensity can effectively penetrate the skull and reach the target brain area, but does not exceed the safety threshold and cause tissue damage.

[0078] When setting the stimulation duration gradient for different sleep cycle regulation targets, it is necessary to consider the matching of the regulation speed of ultrasonic stimulation with the target effect. To promote deep sleep, a longer time parameter is selected, usually in the range of 5-10 minutes. Longer continuous stimulation can more thoroughly regulate the activity of the ventrolateral preoptic area of the anterior hypothalamus and enhance the generation of slow wave sleep; to regulate rapid eye movement sleep, a shorter time parameter is selected, usually in the range of 1-3 minutes. Short-term precise stimulation can avoid excessive interference with the natural process of REM sleep and is sufficient to adjust its timing. In addition, to address the problem of prolonged sleep latency, a medium-length stimulation parameter of 3-5 minutes is set. The determination of the specific duration value also needs to be fine-tuned in combination with individual factors such as the patient's age and weight to ultimately obtain the stimulation duration parameter value.

[0079] Selecting an appropriate waveform type based on the frequency parameter value and the stimulus intensity parameter value is based on the regulation characteristics of different waveforms on neuronal activities. Select continuous wave for slow regulation. The continuous wave provides continuous and stable energy input and is suitable for neural networks that require gradual adjustment, such as the hypothalamic circadian center. Select pulsed wave for fast regulation. The pulsed wave provides high peak energy in a short time and is suitable for neuronal populations that require rapid response, such as the arousal center. Select modulated wave for mixed regulation. The modulated wave combines continuous and periodic changes and is suitable for situations that require both stability and rhythmic adjustment, such as the regulation of transitions between multiple sleep stages. Waveform selection also needs to consider frequency and intensity parameters. The high-frequency and low-intensity scheme is usually paired with a pulsed wave, and the low-frequency and high-intensity scheme is usually paired with a continuous wave, so as to obtain the stimulus waveform parameter value. Combining the frequency parameter value, the stimulus intensity parameter value, the stimulus duration parameter value, and the stimulus waveform parameter value forms an ultrasonic stimulation parameter combination. Set the stimulus focus position according to the rhythm regulation priority. The priority is determined by the main symptoms of sleep disorders: difficulty falling asleep, preferentially select the pineal gland area; difficulty maintaining sleep, preferentially select the hypothalamic area; decreased sleep quality, preferentially select the prefrontal area; abnormal wake-sleep transition, preferentially select the preoptic area. Combine all parameters into a structured data packet containing stimulus program information, thus obtaining an ultrasonic stimulation program.

[0080] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0081] (1) Obtain the tomographic scan image of the user's head, extract the skull structure features and the brain tissue boundaries, and form a three-dimensional brain structure model;

[0082] (2) Mark the boundaries of the pineal gland area, the hypothalamic area, the prefrontal area, and the preoptic area from the three-dimensional brain structure model to obtain a contour map of key brain regions;

[0083] (3) Based on the contour map of key brain regions, calculate the geometric center point of each region and generate an initial set of stimulus point coordinates;

[0084] (4) According to the stimulus focus position in the ultrasonic stimulation program, optimize and adjust the initial set of stimulus point coordinates to generate an accurate set of stimulus point coordinates;

[0085] (5) Measure the thickness and density distribution of the user's skull at each stimulus point position, calculate the attenuation coefficient of the ultrasonic wave penetration path, and form a skull acoustic transmission map;

[0086] (6) Combine the accurate set of stimulus point coordinates and the skull acoustic transmission map, calculate the phase delay matrix and the energy distribution function of the multi-element ultrasonic transducer, and generate spatial positioning data.

[0087] Specifically, to obtain the head tomographic scan image of the user, magnetic resonance imaging (MRI) or computed tomography (CT) technology is usually adopted. The original image data obtained is usually in DICOM format, which contains a series of two-dimensional slice images, and the resolution of each image is 512×512 pixels. These slice images are processed by an image segmentation algorithm to extract the skull structure and the boundary of the brain tissue. In a specific implementation, an automatic segmentation method based on region growing is adopted. First, the gray threshold intervals of the skull, cerebrospinal fluid, gray matter, and white matter are determined, and preliminary segmentation is performed using these thresholds. Then, morphological operations are applied to improve the segmentation results, including opening and closing operations to eliminate noise and fill holes. Finally, these segmentation results are reconstructed in three-dimensional space to form an accurate three-dimensional brain structure model, which is represented by voxels and usually has a resolution of 1×1×1 mm³.

[0088] Import the Talairach or MNI standard brain atlas, and then register the standard atlas with the actual brain model of the user through an elastic deformation algorithm. The registration process adopts the principle of maximizing mutual information, and the spatial correspondence between the two is optimized through iteration to reach the best state. After the registration is completed, the pineal gland region (above the rear end of the third ventricle), hypothalamus region (at the bottom of the third ventricle), prefrontal region (inside the frontal bone), and preoptic area (in the front part of the hypothalamus) can be accurately marked on the user model. Each region is defined by a set of boundary points to form a closed surface, and these closed surfaces together constitute the contour map of the key brain regions. When calculating the geometric center point based on the contour map of the key brain regions, the centroid algorithm is applied to each marked region. First, the region is regarded as a homogeneous body, and then the coordinates of its geometric center point can be calculated by the weighted average of the coordinates of all voxels within the region. The calculation formula is:

[0089] ,

[0090] where, represents the three-dimensional coordinates of the geometric center point of the region, represents all voxels within the region R, represents the coordinates of the voxel v, represents the weight value of the voxel v, which is usually set to 1 or different values are assigned according to tissue density. In this way, each key brain region obtains a geometric center point, and the set of these points forms the initial set of coordinates of the stimulation points.

[0091] When optimizing and adjusting the initial set of stimulation point coordinates according to the stimulation focus position in the ultrasound stimulation protocol, the matching degree between regional functional characteristics and ultrasound parameters needs to be considered. For example, the pineal gland stimulation point is usually slightly offset ventrally to more precisely stimulate the melatonin secretion area, while the hypothalamus stimulation point needs to be accurately positioned at the suprachiasmatic nucleus according to the circadian rhythm regulation target. The optimization and adjustment adopt a correction algorithm based on the functional MRI activation map, using the known functional connection pattern to fine-tune the original geometric center, making the stimulation point more accurately correspond to the key functional area. In addition, the angle and path accessibility of stimulation from outside the skull also need to be considered, avoiding blood vessels and high-density bone areas, and selecting the path with the least ultrasonic penetration resistance. After these optimizations and adjustments, an accurate set of stimulation point coordinates is formed.

[0092] Measuring the thickness and density distribution of the user's skull at each stimulation point position is the basis for calculating the ultrasonic transmission efficiency. The bone density can be directly converted using the attenuation coefficient (Hounsfield unit) in CT data, while for MRI data, it needs to be estimated through the conversion relationship between the T1 image signal intensity and bone density. On the determined ultrasonic incident path, a voxel-by-voxel scan is performed from the outer surface to the inner surface of the skull, recording the density value and path length of each voxel. The calculation formula for the attenuation coefficient of the ultrasonic penetration path in the skull is:

[0093] ;

[0094] where, represents the total attenuation coefficient of the path (dB), Q represents the total number of voxels on the path, represents the material attenuation constant of the q-th voxel , represents the density of the q-th voxel (g / cm³), represents the propagation distance of the ultrasonic wave in the q-th voxel (cm), represents the ultrasonic stimulation frequency (MHz), represents the frequency-dependent index, usually taking values of 1.1 - 1.3. Calculate the attenuation contributions of all voxels on the path, and summarize the path attenuation coefficients at each stimulation point position to form a skull ultrasonic transmission map covering the entire stimulation area. Combining the accurate set of stimulation point coordinates and the skull ultrasonic transmission map, calculate the phase delay matrix and energy distribution function of the multi-element ultrasonic transducer. The multi-element ultrasonic transducer usually consists of dozens of independently controlled piezoelectric elements, and each element can independently adjust the phase and amplitude. The focusing principle is to make the waves coherently superpose at the target position by controlling the phase difference of the waves emitted by each element, forming an energy focus point. The calculation formula for the phase delay matrix is:

[0095] ,

[0096] where, represents the phase delay (in radians) of the g-th element of the transducer relative to the reference element, represents the ultrasonic stimulation frequency (Hz), represents the propagation path length (m) from the g-th element to the target stimulation point e, represents the propagation path length (m) from the reference element to the target point, represents the speed of sound in tissue (m / s), usually 1540 m / s. When calculating the actual phase delay, correction needs to be combined with the skull transmission sonogram to consider the influence of the skull on the speed of sound. The energy distribution function describes the spatial distribution characteristics of the focused ultrasonic field and is usually approximated by a Gaussian model:

[0097] ,

[0098] where, represents the energy density at the spatial point (x, y, z), represents the peak energy density, represents the coordinates of the target stimulation point, represents the focused beam width parameter, which is related to the transducer aperture, frequency, and tissue acoustic impedance. The phase delay matrix and the energy distribution function together constitute the spatial positioning data, providing precise spatial navigation for ultrasonic stimulation.

[0099] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0100] (1) Combine the spatial positioning data with the ultrasonic stimulation scheme to generate a stimulation target description matrix;

[0101] (2) Extract user records similar to the current user's sleep-wake circadian rhythm disorder type from the historical user database to form a reference case set;

[0102] (3) Extract features from the stimulation feedback results in the reference case set, including sleep structure change features, physiological signal change features, and subjective feeling description features, to construct a feedback feature vector library;

[0103] (4) Process the stimulation target description matrix through a three-layer convolutional structure. In the first layer, 32 3×3 convolutional kernels are used to extract low-level features. In the second layer, 64 3×3 convolutional kernels are used to extract intermediate features. In the third layer, 128 3×3 convolutional kernels are used to extract high-level features. After each layer of convolution, a max pooling layer and a batch normalization layer are connected to obtain a multi-dimensional stimulation feature map;

[0104] (5) Perform dimensionality reduction on the multi-dimensional stimulus feature map through two fully-connected layers. The first fully-connected layer has 256 neurons, and the second fully-connected layer has 128 neurons. Use the ReLU activation function to map the high-dimensional features to the sleep parameter evaluation space and form a stimulus effect map.

[0105] (6) Calculate the correlation weights between the stimulus effect map and the feedback feature vector library through cosine similarity, and perform weighted fusion to generate a sleep improvement prediction index, a physiological response prediction value, and a subjective experience expectation score, obtaining prediction feedback data.

[0106] Specifically, combine the spatial localization data with the ultrasound stimulation scheme to form a stimulation description, and predict possible feedback results through deep learning techniques. First, combine the spatial localization data with the ultrasound stimulation scheme to construct a stimulation target description matrix, which is a multi-dimensional data structure containing stimulation parameters, spatial positions, and key brain region information. The specific construction process is to integrate parameters such as frequency, intensity, duration, and waveform in the ultrasound stimulation scheme with coordinate points, phase delays, and energy distributions in the spatial localization data. The rows of the matrix represent different stimulation target points, and the columns represent the parameter attributes of each point, forming a two-dimensional table usually of n×m, where n is the number of stimulation points (usually 4 - 8), and m is the parameter dimension of each point (usually 15 - 20). Extracting similar cases from the historical user database is a process based on user similarity matching. The historical user database contains complete records of users who have received ultrasound stimulation treatment in the past, including basic characteristics, types of sleep disorders, treatment plans, and effect evaluations. Similarity matching uses the weighted nearest neighbor algorithm. First, calculate the distance between the current user and historical users in key features, where the key features include the type of sleep-wake circadian rhythm disorder (weight 0.4), age (weight 0.2), severity of disorder (weight 0.3), and basic physiological parameters (weight 0.1). The distance calculation uses the normalized Euclidean distance to calculate the distance value in the multi-dimensional space after standardizing each feature. Select 10 - 15 user records with the smallest distance from the calculation results to form a reference case set.

[0107] Feature extraction from the stimulus feedback results in the reference case set is a process of converting the original feedback data into structured feature vectors. Sleep structure change features include indicators such as changes in the proportion of deep sleep, changes in the proportion of REM sleep, changes in sleep latency, and changes in the number of nocturnal awakenings; physiological signal change features include indicators such as changes in heart rate variability, changes in respiratory regularity, and changes in body temperature curves; subjective feeling description features extract keywords and emotional tendencies from the user feedback text through natural language processing techniques and quantify them into numerical features. The extracted feature vectors usually contain 20 - 30 dimensions, and each dimension represents a feedback indicator. After being standardized, these feature vectors form a feedback feature vector library, providing a data basis for subsequent similarity calculations.

[0108] Processing the stimulus target description matrix through a convolutional neural network is a deep feature extraction process. First, the two-dimensional matrix is converted into a three-dimensional tensor suitable for convolutional operations, and data augmentation is used to increase sample diversity. The first layer of convolution uses 32 convolutional kernels with a size of 3×3 to perform a sliding window operation on the input data, extracting low-level features such as local spatial relationships and basic parameter patterns. After convolution, the ReLU activation function is applied to introduce non-linearity, and then the feature map size is reduced through a 2×2 max-pooling layer to extract significant features. At the same time, the batch normalization layer is applied to standardize the feature distribution and accelerate the training process. The second layer of convolution uses 64 3×3 convolutional kernels to capture more complex feature combinations and the interaction relationships between parameters, and also undergoes activation, pooling, and normalization processing. The third layer of convolution uses 128 3×3 convolutional kernels to extract highly abstract feature representations, including the pattern characteristics and potential effects of the overall stimulus scheme. These three layers of convolutional structures gradually refine multi-level feature representations from the original data, ultimately forming a multi-dimensional stimulus feature map.

[0109] Reducing the dimension of the multi-dimensional stimulus feature map through a fully connected layer is a process of feature mapping and dimension conversion. The fully connected layer connects all neurons in the previous layer to each neuron in the current layer to achieve the mapping conversion from the feature space to the target space. The first fully connected layer contains 256 neurons, receives the flattened convolutional feature map input, performs a fully connected transformation with the input features through a weight matrix, and then applies the ReLU activation function to introduce non-linearity. The second fully connected layer contains 128 neurons, further compresses the feature dimension, and also applies ReLU activation to output the final low-dimensional feature representation. These two fully connected structures map the high-dimensional convolutional feature space to a 128-dimensional sleep parameter evaluation space, forming a stimulus effect mapping, which encodes the potential relationship from the original stimulus scheme to the expected sleep effect.

[0110] Comparing the stimulus effect mapping with the feedback feature vector library is the core step in generating the prediction result. First, calculate the cosine similarity between the stimulus effect mapping and each vector in the feedback feature vector library. The cosine similarity measures the closeness of the directions of two vectors, and the closer the value is to 1, the more similar they are. Then, weights are assigned according to the similarity values, and cases with higher similarity obtain greater weights. Through the weighted fusion method, calculate the weighted average value to generate the final prediction result, including sleep improvement prediction indicators (such as the percentage increase in deep sleep, the reduction time of sleep latency), physiological response prediction values (such as the improvement degree of heart rate variability), and subjective experience expectation scores (such as sleep quality satisfaction).

[0111] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0112] (1)After the user receives ultrasonic stimulation, monitor the user's sleep cycle data, extract the sleep stage transition frequency, deep sleep proportion, and rapid eye movement sleep duration to form sleep quality assessment indicators;

[0113] (2)Collect the changes in the user's physiological data, including heart rate variability, respiratory rhythm, and body temperature curve, and construct a physiological response record;

[0114] (3)Obtain the user's subjective reports of sleep experiences, including the ease of falling asleep, nocturnal awakening feelings, and morning mental state, and generate a subjective feedback form;

[0115] (4)Integrate the sleep quality assessment indicators, the physiological response record, and the subjective feedback form into actual feedback data;

[0116] (5)By comparing the deviation values of the predicted feedback data and the actual feedback data in each dimension, calculate the feedback difference matrix and generate feedback difference data;

[0117] (6)According to the deviation directions and magnitudes of the parameters in the feedback difference data, precisely adjust the stimulation frequency, stimulation intensity, stimulation duration, stimulation waveform, and stimulation focusing position in the ultrasonic stimulation scheme to obtain a target stimulation scheme.

[0118] Specifically, the personalized plan is continuously improved by accurately comparing the predicted feedback with the actual effect. After the user receives the ultrasonic stimulation, a comprehensive sleep monitoring is required to evaluate the treatment effect. First, the sleep cycle data of the user for 3 - 7 consecutive days is recorded by a portable polysomnography device. These raw data are processed by a sleep staging algorithm, and the sleep is divided into wakefulness, N1, N2, N3, and REM stages according to the characteristics of brain waves, eye movements, and electromyogram activities. The sleep stage transition frequency (i.e., the number of sleep stage changes per hour) is calculated. The normal value is about 4 - 6 times per hour, and a too high value indicates unstable sleep. At the same time, the proportion of deep sleep (N3 stage) in the total sleep time is calculated, which is about 15 - 25% for normal adults, and the duration of rapid eye movement sleep (REM stage), which usually accounts for 20 - 25% of the total sleep time. These three indicators are combined to form a sleep quality assessment index, objectively reflecting the improvement of the sleep structure. Collecting the changes in the user's physiological data is achieved through a variety of biosensors. Heart rate variability is an important indicator for evaluating the function of the autonomic nervous system. The heart rate changes for 24 consecutive hours are recorded by an electrocardiogram monitoring device, and the standard deviation of adjacent R - R intervals (SDNN) and the low - frequency / high - frequency power ratio (LF / HF) are calculated. The respiratory rhythm is recorded by a thoracic and abdominal respiratory belt, and the stability of the respiratory rate and the law of deep and shallow changes are analyzed. During normal sleep, the respiration should be stable and show regular fluctuations. The body temperature curve is recorded by an adhesive skin temperature sensor to record the user's all - day body temperature changes, especially paying attention to the slope of the body temperature drop before going to bed and the time when the lowest body temperature appears during sleep. These are all important indicators of the running state of the biological clock. After these physiological data are processed by digital filtering and artifact removal, a standardized physiological response record is formed, which is used to intuitively reflect the regulatory effect of ultrasonic stimulation on the physiological rhythm.

[0119] Obtaining the user's subjective reported sleep experience is another important dimension for evaluating the treatment effect. User feedback is collected by designing structured sleep diaries and questionnaires. The ease of falling asleep is obtained through the user - reported sleep - onset time and subjective feeling score (1 - 10 points). The nocturnal awakening experience includes the number of awakenings, duration, and degree of wakefulness. The morning mental state is evaluated through multiple dimensions such as freshness, attention concentration, and daytime sleepiness. These subjective data are converted into numerical features through text analysis and quantitative scoring, forming a subjective feedback form, which reflects the perceived improvement of the user's sleep quality.

[0120] When integrating the sleep quality assessment index, physiological response record, and subjective feedback form into actual feedback data, data standardization and structure unification are required. First, all types of data are converted into a unified numerical range (0 - 100 points), and then they are organized into three major categories: physiological indicators, sleep structure indicators, and subjective feeling indicators. Each category of indicators contains multiple specific parameters, forming a structured data table. This integration ensures that feedback data from different dimensions can be compared and processed uniformly, laying a foundation for subsequent comparative analysis with the predicted feedback.

[0121] Calculating the feedback difference matrix by comparing the deviation values of the predicted feedback data and the actual feedback data in each dimension is the core link of the scheme adjustment. The calculation formula of the feedback difference matrix is:

[0122] ,

[0123] where D represents the feedback difference matrix, with the dimension of , b is the number of index categories, r is the number of parameters for each type of index; A represents the actual feedback data matrix, with the same dimension; P represents the predicted feedback data matrix; W represents the index weight matrix, reflecting the importance of different parameters; represents the Hadamard product (element-wise multiplication). Further, normalization processing is introduced to obtain the standardized difference value and eliminate the influence of dimension; finally, through matrix transformation and feature extraction, the difference matrix is converted into structured feedback difference data, clearly showing the deviation direction and degree of each parameter. Precise adjustment of the ultrasonic stimulation scheme according to the feedback difference data is the key step of personalized optimization. The adjustment process follows the gradient response principle, that is, the corresponding parameters are adjusted proportionally according to the deviation direction and amplitude. Specifically, for the deviation of sleep structure, the stimulation frequency and focusing position are mainly adjusted; for the deviation of circadian rhythm, the stimulation intensity and waveform are mainly adjusted; for the deviation of subjective feeling, the stimulation duration and stimulation time are mainly adjusted. For example, when the proportion of deep sleep is lower than expected, reduce the stimulation frequency by 5 - 10% and more precisely locate the focusing position to the ventrolateral preoptic area; when the improvement of heart rate variability is insufficient, appropriately increase the stimulation intensity by 10 - 15% and change to a pulse waveform; when the subjective difficulty of falling asleep does not reach the expected improvement, increase the stimulation duration by 2 - 3 minutes. These adjustments are converted into specific scheme modifications through the parameter mapping matrix to form a more optimized target stimulation scheme.

[0124] The above describes the method for regulating sleep-wake circadian rhythm disorder based on ultrasonic stimulation in the embodiments of the present application. Next, the device for regulating sleep-wake circadian rhythm disorder based on ultrasonic stimulation in the embodiments of the present application will be described. Please refer to Figure 2 , an embodiment of the device for regulating sleep-wake circadian rhythm disorder based on ultrasonic stimulation in the embodiments of the present application includes:

[0125] An acquisition module, configured to acquire the respiratory frequency data, heartbeat data, blood oxygen concentration data, and electromyogram data of the user during sleep, obtain the facial image of the user, record the electroencephalogram data, and construct a sleep cycle feature matrix;

[0126] An analysis module, configured to analyze the sleep stage distribution of the user according to the sleep cycle feature matrix, measure the sleep latency and the number of nocturnal awakenings, and generate a determination result of the sleep-wake circadian rhythm disorder type;

[0127] A generation module, configured to generate a combination of ultrasonic stimulation parameters according to the determination result of the sleep-wake circadian rhythm disorder type, including stimulation frequency, stimulation intensity, stimulation duration, stimulation waveform, and stimulation focusing position, so as to obtain an ultrasonic stimulation plan;

[0128] A stimulation module, configured to determine key brain region stimulation coordinates according to the ultrasonic stimulation plan, including pineal gland region coordinates, hypothalamus region coordinates, prefrontal region coordinates, and preoptic area coordinates, and generate spatial positioning data;

[0129] A prediction module, configured to perform user feedback prediction based on a convolutional neural network according to the ultrasonic stimulation plan through the spatial positioning data, so as to obtain prediction feedback data;

[0130] A correction module, configured to obtain actual feedback data of a user stimulated by the ultrasonic stimulation plan, generate feedback difference data according to the prediction feedback data and the actual feedback data, and correct the ultrasonic stimulation plan according to the feedback difference data to obtain a target stimulation plan.

[0131] Reference Figure 3 , in an embodiment of the present invention, a computer device is further provided. The computer device may be a server, and its internal structure may be as Figure 3 shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0132] Those skilled in the art can understand that Figure 3 the structure shown in

[0133] is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0134] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided in the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0135] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, systems, and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0136] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs, etc., which can store program codes.

[0137] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation, characterized in that: The method for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation includes: Collect the user's breathing rate data, heart rate data, blood oxygen concentration data, and electromyography data during sleep, obtain the user's facial image, record brain wave data, and construct a sleep cycle feature matrix; Analyzing the user's sleep stage distribution based on the sleep cycle characteristic matrix, measuring sleep latency and number of nighttime awakenings, and generating a sleep-wake circadian rhythm disorder type determination result; generating an ultrasound stimulation parameter combination, including stimulation frequency, stimulation intensity, stimulation duration, stimulation waveform, and stimulation focus position, based on the sleep-wake circadian rhythm disorder type determination result, to obtain an ultrasound stimulation plan; According to the ultrasound stimulation scheme, determining the stimulation coordinates of key brain regions, including the coordinates of the pineal region, the hypothalamus region, the prefrontal region, and the preoptic region, and generating spatial positioning data; Through the spatial positioning data, user feedback prediction based on a convolutional neural network is performed according to the ultrasound stimulation scheme to obtain predicted feedback data, including: combining the spatial positioning data with the ultrasound stimulation scheme to generate a stimulation target description matrix; extracting user records with similar sleep-wake circadian rhythm disorder types to the current user from a historical user database to form a reference case set; performing feature extraction on the stimulation feedback results in the reference case set, including sleep structure change features, physiological signal change features and subjective feeling description features, to construct a feedback feature vector library; processing the stimulation target description matrix through a three-layer convolution structure, wherein the first layer uses 32 3×3 convolution kernels to extract low-level features, the second layer uses 64 3×3 convolution kernels to extract mid-level features, and the third layer uses 1 28 3×3 convolution kernels extract high-level features, and each convolution layer is followed by a maximum pooling layer and a batch normalization layer to obtain a multidimensional stimulation feature map. The multidimensional stimulation feature map is subjected to dimensionality reduction processing through two fully connected layers. The first fully connected layer has 256 neurons, and the second fully connected layer has 128 neurons. The high-dimensional features are mapped to the sleep parameter evaluation space using the ReLU activation function to form a stimulation effect map. The stimulation effect map and the feedback feature vector library are calculated using cosine similarity to calculate the correlation weight, and the weighted fusion generates sleep improvement prediction indicators, physiological response prediction values, and subjective experience expectation scores to obtain predicted feedback data. Acquire actual feedback data of users stimulated by the ultrasound stimulation scheme, generate feedback difference data according to the predicted feedback data and the actual feedback data, and modify the ultrasound stimulation scheme according to the feedback difference data to obtain a target stimulation scheme.

2. The method for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation according to claim 1, characterized in that: The method collects the user's respiratory rate data, heart rate data, blood oxygen concentration data, and electromyographic data during sleep, obtains the user's facial image, records brain wave data, and constructs a sleep cycle feature matrix, including: The multi-channel biosensor array collects the user's respiratory rate data, heart rate data, blood oxygen concentration data, and electromyographic data during seven days of continuous sleep to obtain raw physiological data; The infrared imaging device collects the user's facial image once every hour during the user's sleep to obtain a facial image sequence; The user's brain wave data during sleep is recorded by a portable brain wave detection device to obtain brain wave time series data; performing noise elimination and data smoothing processing on the original physiological data to obtain processed physiological data; extracting time domain features and frequency domain features from the processed physiological data to obtain physiological feature values; Extracting facial expression change features from the facial image sequence to obtain facial expression feature values; Separating energy distributions of four wavebands, namely, delta wave, theta wave, alpha wave and beta wave, from the brainwave time series data to obtain brainwave characteristic values; The physiological characteristic values, the facial expression characteristic values and the brain wave characteristic values are integrated in time series to obtain a sleep cycle characteristic matrix.

3. The method for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation according to claim 1, characterized in that: The method of analyzing the user's sleep stage distribution according to the sleep cycle characteristic matrix, measuring sleep latency and the number of nighttime awakenings, and generating a sleep-wake circadian rhythm disorder type determination result includes: Extracting brain wave feature sequences from the sleep cycle feature matrix, dividing the sleep stage boundaries using a feature threshold segmentation method, and obtaining a sleep stage time distribution map; Calculating the time ratio of rapid eye movement sleep to non-rapid eye movement sleep from the sleep stage time distribution diagram to obtain a sleep structure ratio value; The sleep stage time distribution diagram is used to measure the time required from the awake state to the first entry into light sleep to obtain a sleep latency value; Identifying the number of transitions from deep sleep or light sleep to a wakeful state during sleep according to the sleep stage time distribution diagram to obtain the number of nighttime awakenings; Comparing the sleep structure ratio value with the standard sleep ratio range to obtain a structure abnormality index; Obtaining a sleep disorder score by weighted fusion of the sleep latency value, the number of nocturnal awakenings, and the structural abnormality index; According to the sleep disorder score and the temporal pattern of the sleep stage time distribution diagram, and in comparison with the sleep disorder classification standard, a sleep-wake circadian rhythm disorder type determination result is generated.

4. The method for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation according to claim 1, characterized in that: The step of generating an ultrasound stimulation parameter combination based on the sleep-wake circadian rhythm disorder type determination result, including stimulation frequency, stimulation intensity, stimulation duration, stimulation waveform, and stimulation focus position, and obtaining an ultrasound stimulation plan includes: Matching the sleep-wake circadian rhythm disorder type determination result with the ultrasound parameter basic library, extracting the parameter range table of the corresponding type, and obtaining the initial parameter range; According to the initial parameter range, the stimulation frequency is segmented and set to a range of 200-500 kHz for phase-delay disorder, a range of 500-800 kHz for phase-advance disorder, and a range of 800-1200 kHz for irregular disorder, to obtain a frequency parameter value; Based on the frequency parameter value, a safe stimulation intensity range is calculated while taking into account the skull attenuation factor to obtain a stimulation intensity parameter value; According to different sleep cycle regulation targets, the stimulation duration gradient is set, a longer time parameter is selected for deep sleep promotion, and a shorter time parameter is selected for rapid eye movement sleep regulation, and the stimulation duration parameter value is obtained; According to the frequency parameter value and the stimulation intensity parameter value, a suitable waveform type is selected, such as a continuous wave for slow adjustment, a pulse wave for fast adjustment, or a modulated wave for mixed adjustment, to obtain a stimulation waveform parameter value; The frequency parameter value, the stimulation intensity parameter value, the stimulation duration parameter value, and the stimulation waveform parameter value are integrated to form an ultrasound stimulation parameter combination, and the stimulation focus position is set according to the rhythm control priority to obtain an ultrasound stimulation plan.

5. The method for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation according to claim 1, characterized in that: The method of determining the stimulation coordinates of key brain regions according to the ultrasound stimulation scheme, including the coordinates of the pineal region, the hypothalamus region, the prefrontal region, and the preoptic region, and generating spatial positioning data, comprises: Obtain the user's head tomography image, extract the skull structure features and brain tissue boundaries, and form a three-dimensional brain structure model; Marking the pineal region, hypothalamus region, prefrontal lobe region, and preoptic area boundaries in the three-dimensional brain structure model to obtain a contour map of key brain regions; Based on the outline of the key brain areas, the geometric center point of each area is calculated to generate an initial stimulation point coordinate set; Optimizing and adjusting the initial stimulation point coordinate set according to the stimulation focus position in the ultrasound stimulation scheme to generate an accurate stimulation point coordinate set; Measure the thickness and density distribution of the user's skull at each stimulation point, calculate the attenuation coefficient of the ultrasonic penetration path, and form a skull acoustic transmission map; The precise stimulation point coordinate set and the skull acoustic transmission atlas are combined to calculate the phase delay matrix and energy distribution function of the multi-element ultrasonic transducer to generate spatial positioning data.

6. The method for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation according to claim 1, characterized in that: The obtaining of actual feedback data of a user stimulated by the ultrasound stimulation scheme, generating feedback difference data according to the predicted feedback data and the actual feedback data, and modifying the ultrasound stimulation scheme according to the feedback difference data to obtain a target stimulation scheme includes: After the user receives ultrasound stimulation, the user's sleep cycle data is monitored, and the frequency of sleep stage transitions, the proportion of deep sleep, and the duration of rapid eye movement sleep are extracted to form a sleep quality assessment index; Collect changes in the user's physiological data, including heart rate variability, respiratory rhythm, and body temperature curves, and build physiological response records; Obtain users' subjective reports on their sleep experience, including ease of falling asleep, nighttime awakenings, and morning mental state, and generate a subjective feedback form; integrating the sleep quality assessment index, the physiological response record, and the subjective feedback form into actual feedback data; By comparing the deviation values of the predicted feedback data and the actual feedback data in each dimension, a feedback difference matrix is calculated to generate feedback difference data; According to the deviation direction and amplitude of each parameter in the feedback difference data, the stimulation frequency, stimulation intensity, stimulation duration, stimulation waveform and stimulation focus position in the ultrasound stimulation scheme are accurately adjusted to obtain a target stimulation scheme.

7. A sleep-wake circadian rhythm disorder regulating device based on ultrasound stimulation, used to implement the sleep-wake circadian rhythm disorder regulating method based on ultrasound stimulation according to any one of claims 1 to 6, characterized in that: The sleep-wake circadian rhythm disorder regulating device based on ultrasound stimulation comprises: The acquisition module is used to collect the user's respiratory rate data, heart rate data, blood oxygen concentration data, and electromyography data during sleep, obtain the user's facial image, record brain wave data, and construct a sleep cycle feature matrix; An analysis module is used to analyze the user's sleep stage distribution based on the sleep cycle characteristic matrix, measure sleep latency and number of nighttime awakenings, and generate a sleep-wake circadian rhythm disorder type determination result; A generation module is used to generate an ultrasound stimulation parameter combination, including stimulation frequency, stimulation intensity, stimulation duration, stimulation waveform, and stimulation focus position, based on the sleep-wake circadian rhythm disorder type determination result, to obtain an ultrasound stimulation plan; A stimulation module is used to determine the stimulation coordinates of key brain regions according to the ultrasound stimulation scheme, including the coordinates of the pineal region, the hypothalamus region, the prefrontal region, and the preoptic region, and generate spatial positioning data; The prediction module is used to predict user feedback based on a convolutional neural network according to the spatial positioning data and the ultrasonic stimulation scheme to obtain predicted feedback data, including: combining the spatial positioning data with the ultrasonic stimulation scheme to generate a stimulation target description matrix; extracting user records with similar sleep-wake circadian rhythm disorder types to the current user from the historical user database to form a reference case set; extracting features of the stimulation feedback results in the reference case set, including sleep structure change features, physiological signal change features and subjective feeling description features, and constructing a feedback feature vector library; processing the stimulation target description matrix through a three-layer convolution structure, wherein the first layer uses 32 3×3 convolution kernels to extract low-level features, the second layer uses 64 3×3 convolution kernels to extract intermediate features, and the third layer uses 1 28 3×3 convolution kernels extract high-level features, and each convolution layer is followed by a maximum pooling layer and a batch normalization layer to obtain a multidimensional stimulation feature map. The multidimensional stimulation feature map is subjected to dimensionality reduction processing through two fully connected layers. The first fully connected layer has 256 neurons, and the second fully connected layer has 128 neurons. The high-dimensional features are mapped to the sleep parameter evaluation space using the ReLU activation function to form a stimulation effect map. The stimulation effect map and the feedback feature vector library are calculated using cosine similarity to calculate the correlation weight, and the weighted fusion generates sleep improvement prediction indicators, physiological response prediction values, and subjective experience expectation scores to obtain predicted feedback data. The correction module is used to obtain actual feedback data of the user stimulated by the ultrasonic stimulation scheme, generate feedback difference data based on the predicted feedback data and the actual feedback data, and correct the ultrasonic stimulation scheme based on the feedback difference data to obtain a target stimulation scheme.

8. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and is characterized in that when the processor executes the computer program, the method for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the processor is caused to execute the method for regulating sleep-wake circadian rhythm disorders based on ultrasound stimulation according to any one of claims 1 to 6.

Images