Sleep monitoring system, method and device for rhythm regulation

Abstract

The invention provides a sleep monitoring system, method and device for rhythm regulation. Belongs to the technical field of medical electronics and artificial intelligence adjuvant therapy. The method comprises the following steps: positioning multiple physiological signal acquisition areas of a sleep monitoring object, determining EEG signal acquisition points, ECG signal acquisition points and SpO2 signal acquisition points, generating data of the multiple physiological signal acquisition areas, deploying high-timeliness signal acquisition equipment, and constructing a multiple physiological signal synchronous acquisition network. The sleep depth is recognized within 7 seconds or less through the light AI model, the timeliness of sleep state judgment is greatly improved, the atomization administration dosage can be dynamically adjusted in time according to the sleep condition, and precise treatment is achieved.

Description

Technical Field

[0001] This invention proposes a sleep monitoring system, method, and device for rhythm regulation, belonging to the field of medical electronics and artificial intelligence-assisted treatment technology. Background Technology

[0002] In today's fast-paced life, sleep problems are becoming increasingly prominent, posing a significant threat to people's physical and mental health. Traditional sleep therapy techniques have many fundamental shortcomings, making it difficult to meet the needs for efficient, safe, and precise sleep regulation.

[0003] Currently, most mainstream sleep monitoring relies on traditional electroencephalography (EEG) analysis. Its algorithms are complex, with response times as long as 3-6 minutes, making real-time closed-loop medication adjustment impossible and significantly reducing treatment effectiveness. Regarding administration methods, while intravenous injection has a rapid onset of action, it carries high risks such as allergies and vascular damage; oral medications have a slow onset of action and may leave drug residues, affecting treatment efficacy and safety. Furthermore, the lack of safe and efficient non-invasive drug delivery routes limits the widespread adoption and application of sleep therapy.

[0004] Furthermore, existing sleep therapy devices separate monitoring and drug delivery, requiring manual assessment of sleep status and dosage adjustments, which is inefficient and prone to errors. Most devices focus only on the "sleep onset" stage, neglecting long-term rhythm reconstruction and wakefulness management, failing to fundamentally solve sleep problems. Moreover, publicly available sleep therapy solutions mostly focus on smart sleep bracelets, sleep apnea machines, and AI sleep aid apps, failing to deeply integrate elements such as anesthesia depth monitoring, nebulized drug delivery, and closed-loop feedback to build a closed-loop treatment system with autonomous decision-making capabilities. Therefore, developing a novel intelligent sleep rhythm regulation method with ultra-fast response, nebulized drug delivery, closed-loop regulation, rhythm management, and safety self-adaptation capabilities is urgently needed. Summary of the Invention

[0005] This invention provides a sleep monitoring system, method, and apparatus for circadian rhythm regulation, to solve the problems mentioned in the background section above:

[0006] The present invention proposes a sleep monitoring method for rhythm regulation, the method comprising:

[0007] S1. Locate the multi-physiological signal acquisition area for the sleep monitoring subject, determine the EEG signal acquisition point, ECG signal acquisition point and SpO2 signal acquisition point, generate multi-physiological signal acquisition area data, deploy high-efficiency signal acquisition equipment, and build a multi-physiological signal synchronous acquisition network.

[0008] S2. Real-time signal acquisition is performed based on a multi-physiological signal synchronous acquisition network to obtain raw EEG signal data, raw ECG signal data, and raw SpO2 signal data; joint analysis and processing are performed to identify sleep depth data within ≤7 seconds.

[0009] S3. Develop a nebulized drug delivery plan based on sleep depth data, determine the initial nebulized drug delivery dose and nebulized drug delivery frequency, and generate initial nebulized drug delivery parameter data; perform nebulized drug delivery according to the initial nebulized drug delivery parameter data, and continuously collect EEG / ECG / SpO2 signals during the drug delivery process, update sleep depth data in real time, dynamically adjust the nebulized drug delivery dose, and generate dynamically adjusted nebulized drug delivery parameter data.

[0010] S4. Set the treatment cycle, and enter the awake period after the end of each treatment cycle; collect physiological feedback signals of the sleep monitoring subject during the awake period to generate awake period physiological feedback data; optimize and evaluate the dynamically adjusted nebulized drug delivery parameters based on the awake period physiological feedback data to generate optimized nebulized drug delivery parameters data;

[0011] S5. Calculate the weighted rhythm stability index based on the optimized nebulized drug delivery parameters, generate the sleep rhythm stability index, perform risk warning and safety interruption judgment, and generate sleep rhythm regulation treatment data based on the judgment results.

[0012] The sleep monitoring device for circadian rhythm regulation proposed in this invention is characterized by comprising a memory, a processor, and a computer program stored in and executable on the memory, wherein the processor executes the program to implement the sleep monitoring method for circadian rhythm regulation as described above.

[0013] The sleep monitoring system for circadian rhythm regulation proposed in this invention stores a computer program, characterized in that the program is executed by a processor to implement the sleep monitoring method for circadian rhythm regulation as described above.

[0014] The beneficial effects of this invention are as follows: By using a lightweight AI model to identify sleep depth within ≤7 seconds, the timeliness of sleep state assessment is significantly improved, allowing for timely dynamic adjustment of nebulized drug dosage based on sleep patterns, thus achieving precise treatment. Utilizing a multi-physiological signal synchronous acquisition network comprehensively acquires sleep information, reducing the risk of misjudgment due to inaccurate monitoring of a single signal. The 2-hour treatment plus wakefulness period rhythm management process enhances the ability to reconstruct and manage long-term sleep rhythms, contributing to the formation of healthy sleep patterns. The method uses nebulized drug delivery, reducing the risks and inconveniences of traditional intravenous injections and oral medications, and avoiding problems such as allergies, vascular damage, and drug residues. It ensures both the safety and efficiency of sleep therapy while allowing for real-time adjustment based on individual sleep states. Attached Figure Description

[0015] Figure 1 This is a diagram illustrating the steps of the method described in this invention. Detailed Implementation

[0016] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0017] One embodiment of the present invention, such as Figure 1 As shown, a sleep monitoring method for rhythm regulation includes:

[0018] S1. Locate the multi-physiological signal acquisition area for the sleep monitoring subject, determine the EEG (electroencephalogram) signal acquisition point, ECG (electrocardiogram) signal acquisition point, and SpO2 (oxygen saturation) signal acquisition point, and generate multi-physiological signal acquisition area data; deploy high-timeliness signal acquisition equipment based on the multi-physiological signal acquisition area data, and construct a multi-physiological signal synchronous acquisition network;

[0019] S2. Real-time signal acquisition is performed based on a multi-physiological signal synchronous acquisition network to obtain raw EEG signal data, raw ECG signal data, and raw SpO2 signal data; a lightweight AI model is used to jointly analyze and process the raw EEG signal data, raw ECG signal data, and raw SpO2 signal data to identify sleep depth data within ≤7 seconds.

[0020] S3. Develop a nebulized drug delivery plan based on sleep depth data, determine the initial nebulized drug delivery dose and nebulized drug delivery frequency, and generate initial nebulized drug delivery parameter data; perform nebulized drug delivery according to the initial nebulized drug delivery parameter data, and continuously collect EEG / ECG / SpO2 signals during the drug delivery process, and update sleep depth data in real time; dynamically adjust the nebulized drug delivery dose based on the real-time updated sleep depth data, and generate dynamically adjusted nebulized drug delivery parameter data;

[0021] S4. Set a 2-hour treatment cycle, and enter the wake-up period after each treatment cycle; collect physiological feedback signals of the sleep monitoring subject during the wake-up period to generate wake-up period physiological feedback data; optimize and evaluate the dynamically adjusted nebulization drug delivery parameters based on the wake-up period physiological feedback data to generate optimized nebulization drug delivery parameter data;

[0022] S5. Calculate the weighted rhythm stability index based on the optimized nebulization drug delivery parameters to generate a sleep rhythm stability index; perform risk warning and safety interruption judgment based on the sleep rhythm stability index. When the sleep rhythm stability index is lower than the preset safety threshold, automatically interrupt nebulization drug delivery and generate sleep rhythm abnormality warning data; when the sleep rhythm stability index is within the normal range, continue sleep rhythm regulation treatment according to the optimized nebulization drug delivery parameters to generate sleep rhythm regulation treatment data.

[0023] The working principle and effects of the above technical solution are as follows: The solution improves the synchronization and accuracy of physiological signal acquisition during sleep monitoring, reducing errors and interference from single signal acquisition; lightweight AI quickly identifies sleep depth, reducing judgment delay and allowing nebulized drug delivery to adapt to physiological states in a timely manner; dynamically adjusting the drug dosage and periodically optimizing parameters enhances the personalized adaptability of treatment, reducing ineffective drug delivery or inappropriate dosage; the rhythm stability early warning mechanism avoids treatment risks and ensures safe use; it can continuously and dynamically adjust to the monitored subject's sleep state and improve rhythm regulation effects through feedback optimization, thus improving the overall efficiency of sleep monitoring and rhythm regulation.

[0024] In one embodiment of the present invention, S1 includes:

[0025] S11. Conduct an individual physiological baseline assessment of the sleep monitoring subjects, obtain data on age, weight, history of underlying diseases and sleep habits, and generate personalized baseline files for the monitoring subjects;

[0026] S12. Based on personalized baseline profiles and combined with the physiological acquisition principles of EEG, ECG, and SpO2 signals, perform three-dimensional spatial positioning scans on key areas of the monitored subject. The key areas include the head, chest, and fingertips. Determine the optimal anatomical position of each signal acquisition point and generate initial data for multiple physiological signal acquisition areas.

[0027] S13. Perform signal transmission effectiveness tests on each acquisition point in the initial acquisition area data, remove points with signal attenuation risk higher than the preset threshold, recalibrate alternative acquisition points, and generate confirmation data for multiple physiological signal acquisition areas.

[0028] S14. Based on the data collected from the area, select high-efficiency data collection devices that support simultaneous transmission of multiple signals, perform compatibility testing between the devices and the monitored objects, and eliminate compatibility faults.

[0029] S15. Based on the debugged equipment deployment scheme, a multi-physiological signal synchronous acquisition network is built. The multi-physiological signal synchronous acquisition network includes a signal transmission link, a data storage module and a synchronization control unit. The network transmission delay is calibrated to generate a stable multi-physiological signal synchronous acquisition network.

[0030] The working principle and effects of the above technical solution are as follows: The above technical solution improves the accuracy of physiological signal acquisition point positioning, making the acquisition location more closely match the individual characteristics of the monitored object; it reduces the risk of attenuation during signal transmission and enhances the integrity of the acquired signal; the equipment adaptation and debugging eliminates compatibility issues and avoids the interruption of the acquisition network operation; the delay calibration of the synchronous network improves the synchronization of multi-signal transmission and reduces data deviation, which can not only ensure the stable operation of the acquisition network, but also provide high-quality initial data support for subsequent sleep monitoring.

[0031] In one embodiment of the present invention, step S12 includes:

[0032] Physiological structural feature data from personalized baseline profiles are retrieved, and combined with the cortical conduction pattern of EEG signals, the myocardial electrical activity acquisition principle of ECG signals, and the blood oxygen exchange detection mechanism of SpO2 signals, the core physiological targets of each signal acquisition are determined, and data on the correspondence between signals and targets are generated.

[0033] Based on the correspondence data between signals and target points, the key areas of the head, chest and fingertips of the monitored subject are marked and defined on the body surface, and potential collection sub-areas are divided within each area to generate detailed range data of key areas.

[0034] Start the three-dimensional spatial positioning and scanning equipment, and scan each subdivided range layer by layer with an accuracy of 1mm to obtain the three-dimensional coordinates of the body surface and the contour data of the subcutaneous physiological structure of the key areas, and generate a three-dimensional anatomical coordinate dataset.

[0035] The three-dimensional anatomical coordinate dataset is correlated and matched with the signal and target point correspondence data to filter out the coordinate points that fit the core physiological target points, initially lock the candidate acquisition points for each signal, and generate a set of candidate acquisition point coordinates.

[0036] The signal transmission path simulation analysis was performed on the coordinate set of candidate acquisition points to exclude points that affect signal transmission, such as those with excessive subcutaneous fat or bone obstruction, and to retain the coordinates with the best signal transmission efficiency to determine the optimal anatomical position of each signal acquisition point.

[0037] By integrating the three-dimensional coordinates, region, and corresponding signal type information of each optimal anatomical location, initial data for multiple physiological signal acquisition areas are generated.

[0038] The working principle and effects of the above technical solution are as follows: The solution improves the positioning accuracy of physiological signal acquisition points, making the acquisition location more closely aligned with the core physiological targets of each signal; it subdivides key areas and performs precise three-dimensional scanning, reducing surface positioning errors and enhancing the adaptability of acquisition points to subcutaneous physiological structures; it simulates signal transmission paths to eliminate interference factors, avoiding signal transmission obstruction or attenuation, and improving the quality of subsequent acquired signals; it adapts to the individual physiological characteristics of the monitored subject and lays a solid foundation for the simultaneous acquisition of multiple physiological signals, reducing the generation of invalid data due to improper acquisition points.

[0039] In one embodiment of the present invention, S2 includes:

[0040] S21. Start the multi-physiological signal synchronous acquisition network and continuously acquire the original EEG signal data, original ECG signal data and original SpO2 signal data of the monitored object according to the preset sampling frequency of 100Hz, and generate a multi-dimensional original physiological signal dataset.

[0041] S22. Perform adaptive filtering on the original physiological signal dataset to remove power frequency interference, motion artifacts and environmental noise, retain the effective signal frequency band, and generate a preprocessed multi-physiological signal dataset.

[0042] S23. Input the preprocessed data into the lightweight AI model. Through the built-in feature extraction module of the model, extract the sleep cycle features of EEG signal, the heart rate variability features of ECG signal, and the blood oxygen fluctuation features of SpO2 signal to generate a multimodal physiological feature set.

[0043] S24. The AI ​​model performs fusion analysis on the multimodal physiological feature set and combines it with the sleep staging algorithm to complete the classification and identification of sleep depth within ≤7 seconds, generating preliminary sleep depth data, which includes light sleep, deep sleep and REM sleep.

[0044] S25. Perform temporal continuity verification on the preliminary sleep depth data, correct abnormal jump data points, and generate the final confirmed sleep depth data.

[0045] The working principle and effects of the above technical solution are as follows: adaptive filtering reduces the impact of interference and noise on the effective signal, enhances data purity, and improves the density and completeness of physiological signal acquisition; the lightweight AI model integrates multimodal feature analysis, which significantly shortens the time for sleep depth recognition and improves judgment efficiency; temporal coherence verification corrects abnormal data points, reduces the probability of misjudgment, and avoids inappropriate subsequent treatment due to data deviation; it can quickly capture the dynamic changes of physiological signals of the monitored subject and ensure the accuracy of sleep depth recognition, providing reliable data support for subsequent personalized drug delivery adjustments.

[0046] In one embodiment of the present invention, S3 includes:

[0047] S31. Retrieve medication contraindication information from the personalized baseline file, combine it with the rhythm regulation needs corresponding to sleep depth data, refer to clinical nebulized drug administration guidelines, formulate an initial nebulized drug administration plan, determine the core drug components, initial nebulized drug dosage and nebulized drug administration frequency, and generate a draft of initial nebulized drug administration parameters.

[0048] S32. Perform dose safety verification on the draft initial nebulization drug delivery parameters to ensure that the single dose and the cumulative daily dose do not exceed the safety threshold, correct the parameters that exceed the threshold, and generate the initial nebulization drug delivery parameter data.

[0049] S33. Start the nebulization drug delivery device based on the initial nebulization drug delivery parameter data, maintain the operation of the multi-physiological signal synchronous acquisition network during the drug delivery process, collect and update EEG, ECG, and SpO2 signal data in real time, and generate dynamic physiological signal stream;

[0050] S34. Based on dynamic physiological signal flow, calculate the rate of change and fluctuation amplitude of sleep depth data, establish a correlation model between signal change and drug dosage, dynamically correct the nebulized drug dosage according to the principle of gradient adjustment and small incremental increase, and generate dynamically adjusted nebulized drug delivery parameter data.

[0051] S35. Perform real-time compliance checks on the dynamically adjusted parameter data to ensure that the adjustment range is within the clinical safety range, and output the final dynamically adjusted nebulization drug delivery parameter data.

[0052] The working principle and effects of the above technical solution are as follows: Initial plans are developed by combining individual baselines and sleep states, reducing the incompatibility of general plans and improving the personalized suitability of nebulized drug delivery plans; dose safety verification and compliance checks reduce medication risks and avoid adverse consequences caused by overdosing; real-time acquisition of physiological signals and dynamic adjustment of dosage enhance the synchronization between drug delivery and physiological state, reducing ineffective drug delivery or inappropriate dosage; it ensures drug safety while precisely optimizing dosage based on changes in sleep depth, improving the therapeutic effect of rhythm regulation and making the drug delivery process more closely aligned with the real-time needs of the monitored individual.

[0053] In one embodiment of the present invention, S33 includes:

[0054] The device operating parameters, including atomization pressure, drug atomization particle size, and drug delivery start / stop sequence, are retrieved from the initial nebulization drug delivery parameter data. The parameters are then imported into the nebulization drug delivery device control system to complete the device parameter initialization configuration and generate a device start-up ready signal.

[0055] Based on the device startup and readiness signal, a collaborative operation command is sent to the multi-physiological signal synchronous acquisition network to activate the signal transmission link, data storage module and synchronization control unit in the network, confirm the timing synchronization between the network and the drug delivery device, and generate network collaborative operation status data.

[0056] Based on the real-time requirements of sleep monitoring, the signal acquisition update frequency (50Hz) and data transmission delay threshold (≤15ms) are set to generate real-time configuration data for signal acquisition.

[0057] The nebulized drug delivery device is activated to perform drug delivery. At the same time, according to the real-time configuration data, the real-time changes of EEG, ECG, and SpO2 signals are continuously captured through the acquisition points of the multi-physiological signal synchronous acquisition network to generate a set of real-time signal segments frame by frame.

[0058] The time-series alignment process is performed on the frame-by-frame real-time signal segment set to remove invalid segments with excessive transmission delay, and the valid signal data is spliced ​​together in timestamp order to generate a dynamic physiological signal stream with a uniform format.

[0059] The working principle and effects of the above technical solution are as follows: The above technical solution improves the synergy between nebulized drug delivery and physiological signal acquisition, allowing the drug delivery operation and signal acquisition to be synchronized in real time; setting clear acquisition frequency and delay standards enhances the real-time performance of signal acquisition and reduces the problem of data transmission lag; time-series alignment processing removes invalid segments, improving the integrity and accuracy of dynamic physiological signal flow and avoiding the impact of interference data on subsequent analysis; it can not only ensure the stable execution of the drug delivery process, but also provide high-quality real-time data support for dynamic dose adjustment, making the connection of the entire treatment process smoother.

[0060] In one embodiment of the present invention, S34 includes:

[0061] Feature extraction is performed on dynamic physiological signal streams to separate the sleep cycle rhythm features of EEG signals, the heart rate variability features of ECG signals, and the blood oxygen homeostasis features of SpO2 signals, generating a multi-dimensional physiological signal feature set.

[0062] Based on feature sets and historical sleep depth data, a sliding window algorithm is used to calculate the rate of change (unit: levels / minute) and fluctuation amplitude (unit: levels) of sleep depth per unit time, generating dynamic sleep depth change index data;

[0063] By defining signal change indicators (rate of change, fluctuation amplitude) as input variables and dosage adjustment as output variables, a nonlinear regression model framework is constructed to generate a preliminary dose-related model.

[0064] Import the constraint parameters corresponding to the gradient adjustment and small incremental principle. The constraint parameters include a single maximum adjustment amplitude ≤10% and an adjacent adjustment interval ≥5 minutes. Configure the constraint conditions for the prototype model to generate a signal and dose correlation model.

[0065] The dynamic change index data of sleep depth is input into the correlation model to calculate the dose adjustment value adapted to the current physiological state and generate nebulized drug delivery dose correction data. The dose field in the initial nebulized drug delivery parameter data is updated with the dose correction data, while retaining reasonable parameters such as the original frequency, to generate dynamically adjusted nebulized drug delivery parameter data.

[0066] The working principle and effects of the above technical solution are as follows: The solution improves the comprehensiveness of physiological signal feature extraction, allowing dosage adjustments to align with multi-dimensional physiological state changes; it accurately calculates the rate and amplitude of sleep depth changes, enhancing the targeted nature of dynamic adjustments and reducing delayed adjustments; the constraint parameters standardize the adjustment amplitude and interval, reducing the risk of excessive dosage fluctuations and preventing improper adjustments from affecting treatment efficacy; it achieves precise dosage adaptation through a correlation model while ensuring a stable and safe adjustment process, ensuring that nebulized drug delivery always aligns with the monitored subject's real-time sleep state, thus improving the effectiveness of rhythm regulation.

[0067] In one embodiment of the present invention, step S4 includes:

[0068] S41. Based on the sleep depth fluctuation pattern of the previous stage, set a fixed 2-hour treatment cycle, and set a monitoring node every 15 minutes within the cycle to generate a treatment cycle monitoring plan.

[0069] S42. After each treatment cycle, a wake-up period initiation command is triggered, and the monitored subject is guided into a light sleep state through gentle sound and light stimulation. The light sleep state does not affect subsequent sleep, and a wake-up period initiation signal is generated.

[0070] S43. During the wakefulness period, subjective sleep experience feedback data and objective physiological feedback signals of the monitored subject are collected simultaneously to generate a comprehensive physiological feedback dataset during the wakefulness period. The subjective sleep experience feedback data includes whether the subject is comfortable and whether there are any discomfort symptoms. The objective physiological feedback signals include heart rate, blood oxygen saturation and simple electroencephalogram (EEG) activity.

[0071] S44. Based on the comprehensive physiological feedback dataset, establish a quantitative model for the effect of parameter adjustment, compare the improvement in sleep depth and the stability of physiological indicators before and after dynamic adjustment, and conduct multi-dimensional optimization and evaluation of the nebulized drug delivery parameter data after dynamic adjustment.

[0072] S45. Based on the evaluation results, correct any unreasonable parameters in the dosage and frequency of drug administration, generate optimized nebulization drug administration parameter data, and verify the applicability of these parameters in subsequent cycles.

[0073] The working principle and effects of the above technical solution are as follows: A 2-hour cycle combined with monitoring nodes allows assessments to better align with treatment patterns. Gentle audio-visual stimulation guides light awakening, avoiding the interference of deep arousal on subsequent sleep, reducing discomfort for the monitored individual, and improving the targeted nature of nebulized drug delivery parameter optimization. Combining subjective feelings with objective physiological signals enhances the comprehensiveness of feedback data and reduces the bias of a single assessment dimension. A quantitative model evaluates parameter effects from multiple dimensions, improving the accuracy of parameter correction and avoiding blind adjustments. This allows the drug delivery regimen to continuously adapt to changes in sleep state and ensures parameter applicability through periodic validation, further enhancing the long-term effects of rhythm regulation.

[0074] In one embodiment of the present invention, step S5 includes:

[0075] S51. Based on the rhythm consistency of EEG signals, the heart rate stability of ECG signals, and the steady-state value of blood oxygen saturation of SpO2 signals, determine the weighting factors of each indicator in the rhythm stability assessment. The weighting factors include EEG weight 0.4, ECG weight 0.3, and SpO2 weight 0.3, and generate a weighting configuration scheme.

[0076] S52. Based on the weighting scheme, the real-time physiological signals under the action of the optimized nebulized drug delivery parameters are weighted and calculated to generate a sleep rhythm stability index.

[0077] S53. Perform dynamic threshold calibration on the sleep rhythm stability index, combine it with the individualized baseline profile of the monitored individual to determine the appropriate safety threshold range for the individual, and generate an individualized safety threshold standard.

[0078] S54. The real-time calculated sleep rhythm stability index is compared with the personalized safety threshold standard. If the index is lower than the safety threshold, the nebulized drug delivery interruption command is automatically triggered, and sleep rhythm abnormality warning data is generated at the same time. The sleep rhythm abnormality warning data includes the abnormality type, possible cause and emergency treatment suggestions.

[0079] S55. If the index is within the normal range, continue to perform sleep rhythm regulation treatment according to the optimized nebulized drug delivery parameters, record the changes in physiological indicators and parameter execution in real time during the treatment process, generate complete sleep rhythm regulation treatment data, and synchronize it to the remote monitoring platform.

[0080] The working principle and effects of the above technical solution are as follows: multi-index weighted calculation makes the assessment more in line with the actual physiological state, improving the comprehensiveness and accuracy of sleep rhythm stability assessment; dynamic calibration of personalized safety thresholds enhances the adaptability of judgment and reduces misjudgment caused by general standards; automatic interruption of drug administration and generation of warning when the index is abnormal, avoiding the expansion of risk and reducing treatment safety hazards; real-time recording of treatment data and synchronization with the remote platform make the monitoring process traceable and improve management convenience; it can not only ensure treatment safety throughout the process, but also make remote monitoring more efficient, further strengthening the reliability of rhythm regulation therapy.

[0081] An embodiment of the present invention provides a sleep monitoring device for circadian rhythm regulation, characterized in that it includes a memory, a processor, and a computer program stored in and executable on the memory, wherein the processor executes the program to implement the sleep monitoring method for circadian rhythm regulation as described above.

[0082] One embodiment of the present invention provides a sleep monitoring system for circadian rhythm regulation, wherein a computer program is stored thereon, characterized in that the program is executed by a processor to implement the sleep monitoring method for circadian rhythm regulation as described in any of the above.

[0083] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A sleep monitoring method for circadian regulation, characterized in that, The method comprises: S1, positioning the multi-physiological signal acquisition area of the sleep monitoring object, determining the EEG signal acquisition point, the ECG signal acquisition point and the SpO2 signal acquisition point, generating the multi-physiological signal acquisition area data, deploying the high-time-efficiency signal acquisition equipment, and constructing the multi-physiological signal synchronous acquisition network; S2, based on the multi-physiological signal synchronous acquisition network, acquiring the original EEG signal data, the original ECG signal data and the original SpO2 signal data through real-time signal acquisition; performing joint analysis and processing to identify the sleep depth data within ≤7 seconds; S3, according to the sleep depth data, formulating the atomization drug delivery scheme, determining the initial atomization drug delivery dose and the atomization drug delivery frequency, generating the initial atomization drug delivery parameter data; performing atomization drug delivery according to the initial atomization drug delivery parameter data, continuously acquiring the EEG / ECG / SpO2 signal during the drug delivery process, updating the sleep depth data in real time, dynamically adjusting the atomization drug delivery dose, and generating the dynamically adjusted atomization drug delivery parameter data; S4, setting a treatment cycle, entering a wake-up period after each treatment cycle; acquiring the physiological feedback signal of the sleep monitoring object in the wake-up period to generate the wake-up period physiological feedback data; optimizing and evaluating the dynamically adjusted atomization drug delivery parameter data through the wake-up period physiological feedback data to generate the optimized atomization drug delivery parameter data; S5, calculating the weighted rhythm stability index according to the optimized atomization drug delivery parameter data to generate the sleep rhythm stability index, performing risk warning and safety interruption judgment, and generating the sleep rhythm regulation treatment data according to the judgment result.

2. The sleep monitoring method for rhythm regulation according to claim 1, characterized in that, The S1 comprises: S11, performing individual physiological characteristic baseline evaluation on the sleep monitoring object to acquire age, weight, disease history and sleep habit data, and generating the individualized baseline profile of the monitoring object; S12, based on the individualized baseline profile, combining the physiological acquisition principles of EEG, ECG and SpO2 signals, performing three-dimensional space positioning scanning on the key area of the monitoring object to determine the optimal anatomic position of each signal acquisition point, and generating the initial data of the multi-physiological signal acquisition area; S13, performing signal conduction effectiveness test on each acquisition point in the initial acquisition area data, eliminating the points with signal attenuation risk higher than the preset threshold, recalibrating the alternative acquisition points, and generating the confirmed data of the multi-physiological signal acquisition area; S14, according to the acquisition area confirmation data, screening the high-time-efficiency acquisition equipment supporting multi-signal synchronous transmission, performing adaptability debugging of the equipment and the monitoring object, and excluding compatibility faults; S15, based on the equipment deployment scheme after debugging, building the multi-physiological signal synchronous acquisition network, calibrating the network transmission delay, and generating the stable multi-physiological signal synchronous acquisition network.

3. The sleep monitoring method for rhythm regulation according to claim 2, characterized in that, The S12 comprises: Retrieving the physiological structure characteristic data in the individualized baseline profile, combining the brain cortex conduction rule of EEG signal, the myocardial electrical activity acquisition principle of ECG signal and the blood oxygen exchange detection mechanism of SpO2 signal, and generating the signal and target point correspondence data; Based on the signal and target correspondence data, the key areas of the head, chest and fingertips of the monitored object are marked and the range is defined, the potential collection sub-regions in each region are divided, and the key area subdivision range data is generated; Start the three-dimensional space positioning scanning device, and scan each subdivision range layer by layer with 1mm precision to obtain the body surface three-dimensional coordinates and subcutaneous physiological structure contour data of the key area, and generate three-dimensional anatomical coordinate data set; Correlate and match the three-dimensional anatomical coordinate data set with the signal and target correspondence data, filter out the coordinate points matching the core physiological target, preliminarily lock the candidate collection points of each signal, and generate the candidate collection point coordinate set; Simulate and analyze the signal conduction path of the candidate collection point coordinate set to determine the optimal anatomical position of each signal collection point; Integrate the three-dimensional coordinates of each optimal anatomical position, the corresponding signal type information and the region information to generate the initial data of the multi-physiological signal collection region.

4. The sleep monitoring method for rhythm regulation according to claim 1, characterized in that, The S2 comprises: S21, start the multi-physiological signal synchronous collection network, continuously collect the original EEG signal data, original ECG signal data and original SpO2 signal data of the monitored object according to the preset 100Hz sampling frequency, and generate the multi-dimensional original physiological signal data set; S22, perform adaptive filtering processing on the original physiological signal data set to generate the pre-processed multi-physiological signal data set; S23, input the pre-processed data into the lightweight AI model, extract the sleep cycle feature of the EEG signal, the heart rate variability feature of the ECG signal and the blood oxygen fluctuation feature of the SpO2 signal through the feature extraction module built in the model, and generate the multi-modal physiological feature set; S24, the AI model performs fusion analysis on the multi-modal physiological feature set, combines the sleep staging algorithm, and completes the classification and identification of sleep depth within ≤7 seconds to generate the preliminary sleep depth data; S25, perform time sequence coherence verification on the preliminary sleep depth data, correct abnormal jump data points, and generate the final confirmed sleep depth data.

5. The sleep monitoring method for circadian rhythm modulation of claim 1, wherein, The S3 comprises: S31, retrieve the drug contraindication information in the individualized baseline file, combine the rhythm adjustment demand corresponding to the sleep depth data, refer to the clinical atomization drug administration guideline, formulate the initial atomization drug administration scheme, determine the core drug component, the initial atomization drug administration dose and the atomization drug administration frequency, and generate the initial atomization drug administration parameter draft; S32, perform dose safety verification on the initial atomization drug administration parameter draft to generate the initial atomization drug administration parameter data; S33, start the atomization drug administration device based on the initial atomization drug administration parameter data, keep the multi-physiological signal synchronous collection network running during the drug administration process, collect and update the EEG, ECG and SpO2 signal data in real time, and generate the dynamic physiological signal stream; S34, based on the dynamic physiological signal stream, calculate the change rate and fluctuation amplitude of the sleep depth data, establish the correlation model of signal change and drug dose, dynamically correct the atomization drug administration dose, and generate the dynamically adjusted atomization drug administration parameter data; S35, perform real-time compliance check on the dynamically adjusted parameter data to ensure that the adjustment amplitude is within the clinical safety range, and output the final dynamic adjustment atomization drug administration parameter data.

6. The sleep monitoring method for rhythm regulation according to claim 5, characterized in that, The S33 comprises: The device operation parameter in the initial atomization administration parameter data is called, the parameter is introduced into the atomization administration device control system, the device parameter initialization configuration is completed, and a device start ready signal is generated; Based on the device start ready signal, a cooperative operation instruction is sent to the multi-physiological signal synchronous acquisition network, the signal transmission link, data temporary storage module and synchronous control unit in the network are activated, the time sequence synchronization of the network and the administration device is confirmed, and network cooperative operation state data is generated; According to the real-time requirement of sleep monitoring, the signal acquisition update frequency and data transmission delay threshold are set, and signal acquisition real-time configuration data is generated; The atomization administration device is started to perform administration operation, and at the same time, according to the real-time configuration data, the real-time changes of EEG, ECG and SpO2 signals are continuously captured through each acquisition point of the multi-physiological signal synchronous acquisition network, and a frame-by-frame real-time signal segment set is generated; The frame-by-frame real-time signal segment set is subjected to time sequence alignment processing, invalid segments with transmission delay exceeding the threshold are removed, effective signal data is spliced in the timestamp order, and a dynamic physiological signal stream with uniform format is generated.

7. The sleep monitoring method for rhythm regulation according to claim 1, characterized in that, The S4 comprises: S41, according to the sleep depth fluctuation rule of the previous stage, a 2-hour fixed treatment cycle is set, a monitoring node is set every 15 minutes in the cycle, and a treatment cycle monitoring plan is generated; S42, after each treatment cycle, an awakening period start instruction is triggered, the monitoring object is guided into a light awakening state through mild sound and light stimulation, and an awakening period start signal is generated; S43, in the awakening period, the subjective sleep feeling feedback data and the objective physiological feedback signal of the monitoring object are synchronously acquired, and an awakening period physiological feedback comprehensive data set is generated; S44, based on the physiological feedback comprehensive data set, a parameter adjustment effect quantitative model is established, the sleep depth improvement degree and the physiological index stability before and after dynamic adjustment are compared, and the multi-dimensional optimization evaluation of the atomization administration parameter data after dynamic adjustment is carried out; S45, according to the evaluation result, unreasonable parameters in the administration dose and frequency are corrected, optimized atomization administration parameter data is generated, and the applicability of the parameters in the subsequent cycle is verified.

8. The sleep monitoring method for rhythm regulation according to claim 1, characterized in that, The S5 comprises: S51, based on the rhythm consistency of the EEG signal, the heart rate stability of the ECG signal and the blood oxygen saturation steady-state value of the SpO2 signal, the weight factor of each index in the rhythm stability evaluation is determined, and a weight configuration scheme is generated; S52, according to the weight configuration scheme, the real-time physiological signal under the optimized atomization administration parameter data is subjected to weighted calculation, and a sleep rhythm stability index is generated; S53, the sleep rhythm stability index is subjected to dynamic threshold calibration, the individualized baseline file of the monitoring object is combined to determine the safety threshold range suitable for the object, and an individualized safety threshold standard is generated; S54, the sleep rhythm stability index calculated in real time is compared with the individualized safety threshold standard, if the index is lower than the safety threshold, an atomization administration interruption instruction is automatically triggered, and sleep rhythm abnormality early warning data is generated. S55, if the index is in the normal range, continue to perform the sleep rhythm adjustment treatment according to the optimized atomization administration parameter data, record the physiological index changes and parameter execution in the treatment process in real time, generate complete sleep rhythm adjustment treatment data, and synchronize to the remote monitoring platform.

9. Sleep monitoring device for circadian regulation, characterized in that, A computer program product comprising a memory, a processor and a computer program stored on the memory and loadable into the memory, the processor being configured to execute the program to implement the sleep monitoring method for rhythm adjustment according to any one of claims 1 to 8.

10. A sleep monitoring system for circadian regulation, having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the sleep monitoring method for rhythm adjustment according to any one of claims 1 to 8.

Images