A sleep state recognition and intervention system

Abstract

This application discloses a sleep state recognition and intervention system, belonging to the field of smart healthcare. In this system, a multimodal data acquisition module acquires the user's physiological data, electroencephalogram (EEG) signal data, and assessment scale data as multimodal data. An intelligent decision-making module fuses and analyzes the multimodal data to calculate a comprehensive psychological sleep stress index. Based on judgments across five dimensions—attention state, stress level, sleep problems, depression level, and anxiety level—it applies preset dynamic adjustment coefficients to calculate scores for each meridian, identifying the core symptom meridians. A collaborative intervention module matches the comprehensive psychological sleep stress index and the core symptom meridians with corresponding target intervention scenarios, and controls the semi-enclosed capsule and smart headgear to perform physical linkage. It also controls the transcranial magnetic stimulation device and the hydrogen-oxygen intervention device to perform physical intervention operations according to a preset collaborative intervention protocol, improving the accuracy of sleep state recognition and the efficacy of clinical intervention.

Description

Technical Field

[0001] This application relates to the field of smart healthcare technology, and in particular to a sleep state recognition and intervention system. Background Technology

[0002] Sleep disorders are a prevalent health problem worldwide. Current sleep assessment and intervention techniques suffer from the following significant technical limitations.

[0003] First, the fragmented nature of multi-source, heterogeneous data lacks a fusion and computation mechanism. Existing assessment methods either rely on subjective scale completion, which is susceptible to memory bias and data distortion, or on single objective physiological devices such as polysomnography (PSG), which are expensive and cannot reflect psychological and cognitive states. Physiological data, EEG signals, and psychological scale data operate independently, lacking a unified data normalization and multi-dimensional weighted fusion algorithm, resulting in the system's inability to output a comprehensive and accurate overall stress index.

[0004] Secondly, there are strict data barriers between the assessment and intervention ends, lacking a dynamic parameter configuration mechanism. Existing intervention devices (such as simple transcranial magnetic stimulation devices or music relaxation chairs) typically use fixed factory treatment parameters. There is no data communication bus between the assessment and intervention devices, and the computer equipment cannot dynamically generate and issue personalized hardware control commands based on the multimodal fusion results collected at the front end, resulting in a mismatch between intervention parameters and the user's real-time pathological state.

[0005] Finally, multimodal physical intervention devices lack spatial linkage and temporal coordinated control. Existing integrated treatment cabins often simply stack multiple physical modules, with each module operating independently. The system lacks automated spatial positioning control for mechanical structures such as the semi-enclosed cabin and smart headgear, and also lacks temporal coordinated protocols for multiple energy output devices such as transcranial magnetic stimulation, hydrogen-oxygen therapy, and somatosensory music. This results in physical intervention remaining at the level of superimposing single methods, failing to form a closed-loop intervention path of "environment creation - system regulation - central control," thus limiting clinical efficacy. Summary of the Invention

[0006] To address the technical problems in existing technologies, such as fragmented multi-source data, lack of data interoperability between assessment and intervention devices leading to parameter rigidity, and lack of spatial linkage and temporal coordinated control of multiple physical intervention devices, this application provides a sleep state recognition and intervention system.

[0007] This application is achieved through the following technical solution.

[0008] A sleep state recognition and intervention system, comprising:

[0009] The multimodal data acquisition module is used to acquire the user's physiological data, electroencephalogram (EEG) signal data, and assessment scale data as multimodal data. The physiological data includes the user's blood oxygen, blood pressure, heart rate, and heart rate variability data. The EEG signal data includes delta wave, theta wave, alpha wave, beta wave, and gamma wave data. The assessment scale data includes the user's scores on the Zung Self-Rating Depression Scale, the Generalized Anxiety Scale, and a customized sleep questionnaire.

[0010] The intelligent decision-making module, connected to the multimodal data acquisition module, is used to perform fusion analysis on multimodal data, calculate a comprehensive psychological sleep stress index, and, based on the judgment conclusions of five dimensions—attention state, stress status, sleep problems, depression level, and anxiety level—calculate the scores of each meridian using preset dynamic adjustment coefficients to determine the core symptom meridians; and

[0011] The collaborative intervention module, connected to the intelligent decision-making module, is used to match the target intervention scenario according to the comprehensive psychological sleep stress index and the core symptom meridians, and to control the semi-enclosed cabin and intelligent headgear to perform physical linkage, and to control the transcranial magnetic stimulation device and hydrogen-oxygen intervention device to perform physical intervention operations according to the preset collaborative intervention protocol.

[0012] Optionally, the intelligent decision-making module performs fusion analysis on multimodal data to calculate a comprehensive psychological sleep stress index, specifically including:

[0013] Using an improved eSense digital measurement algorithm, EEG signal data is converted into a dual-dimensional quantitative index of focus and relaxation, and an EEG imbalance index is further calculated. ;

[0014] The arithmetic sum of the scores for all items in the customized sleep questionnaire is taken as the actual total score of the sleep questionnaire. And further calculate the sleep scale score. ;

[0015] Statistical analysis of the total score of the Generalized Anxiety Scale And further calculate anxiety scores. ;

[0016] Calculate the total score of the Zung Self-Rating Depression Scale. And further converted into SDS standard scores. As a depression score ;

[0017] Based on the collected heart rate and heart rate variability data, a stepwise scoring system was used to determine the heart rate score by comparing it with the stress threshold of the autonomic nervous system. The heart rate score was then mapped to a percentage scale as the heart rate abnormality score. ;

[0018] A stepwise scoring system is used to determine the blood oxygen score based on the magnitude of the decrease in blood oxygen concentration and the risk of sleep apnea. This score is then mapped to a percentage scale as the abnormal blood oxygen score. ;

[0019] Blood pressure zones are divided into systolic, diastolic, and normal ranges, and scores are assigned accordingly to determine blood pressure scores. These scores are then mapped to a percentage scale to serve as blood pressure abnormality scores. ;

[0020] Using formula Calculate attention deficit score ;in To focus on quantifiable indices;

[0021] Using formula Calculate the comprehensive psychological sleep stress index .

[0022] Optionally, the brainwave imbalance index The calculation formula is: ;in To relax the quantification index.

[0023] Optionally, the sleep scale score The calculation formula is: .

[0024] Optionally, the anxiety score The calculation formula is: .

[0025] Optionally, the depression score The calculation formula is: ;in This represents the "rounding to the nearest integer" function.

[0026] Optionally, the intelligent decision-making module, based on the judgment conclusions of five dimensions—attention state, stress state, sleep problems, depression level, and anxiety level—applies a preset dynamic adjustment coefficient to calculate the score of each meridian and determine the core symptom meridians, specifically including:

[0027] The judgment conclusions of the five dimensions are obtained. Taking the moderate judgment conclusion as the baseline, the standard scores of the twelve meridians are weighted and calculated using adjustment coefficients:

[0028] Apply the ±25% rule: if the conclusion of any dimension is one level higher or lower than the average, the corresponding meridian standard score is multiplied by 1.25 or 0.75.

[0029] Apply the ±50% rule: If the anxiety level dimension is determined to be no anxiety, the corresponding meridian standard score is multiplied by 0.5;

[0030] The scores for the same meridian calculated from the five dimensions are added together and averaged to obtain a preliminary comprehensive score;

[0031] We introduce the core meridians corresponding to the psychological diagnosis and treatment classification, and multiply the initial comprehensive score by 0.75 to correct the classification.

[0032] The final scores of all meridians are sorted, and the meridian with the lowest score is identified as the core symptom meridian.

[0033] Optionally, the collaborative intervention module controls the semi-enclosed cabin and the smart helmet to perform physical linkage, specifically including:

[0034] The semi-enclosed cabin is controlled to automatically close to form a closed intervention space;

[0035] The intelligent headgear is controlled to be lowered non-contactly above the user's head to accurately position the transcranial magnetic stimulation unit and / or bio-light wave emission unit;

[0036] Adjust the reclining mechanism inside the semi-enclosed cabin to a 15° slightly tilted flat position.

[0037] Optionally, the timing logic defined by the preset collaborative intervention protocol includes:

[0038] First, the cabin ambient lighting and motion-sensing music equipment are activated, converting the 16-150Hz mid-low frequency signal into physical vibrations that are transmitted to the user through bone conduction.

[0039] The hydrogen-oxygen intervention device is activated simultaneously or subsequently to adjust the whole-body system according to the preset hydrogen concentration and flow parameters of the matched target intervention scenario;

[0040] Finally, the transcranial magnetic stimulation device is used to perform central nervous system modulation according to the preset stimulation frequency, intensity, and brain region target parameters of the matched target intervention scenario.

[0041] Optionally, the intelligent decision-making module is also used to execute physiological-psychological cross-validation logic:

[0042] When the assessment scale data shows severe insomnia, but the EEG signal monitoring shows that the proportion of delta wave energy and heart rate variability are within the normal range, it is judged as subjective insomnia, and an intervention scenario containing cognitive behavioral therapy audio is automatically matched.

[0043] When the assessment scale data shows no snoring symptoms, but the blood oxygen saturation monitoring shows that the nighttime oxygen reduction index is higher than the preset threshold, it is determined to be concealed sleep apnea. The system automatically matches a deep sleep scenario and forcibly increases the flow rate parameter of the hydrogen-oxygen intervention device to 3.0L / min, while generating a polysomnography re-examination prompt.

[0044] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0045] This application provides a sleep state recognition and intervention system. On the one hand, by constructing a unified normalization and multidimensional weighted fusion algorithm, it cross-verifies and fuses objective physiological data, EEG characteristics, and subjective scale data to output the Comprehensive Psychological Sleep Stress Index (TPSSI). This effectively overcomes the bias and lag in assessment caused by a single data source, breaks down data silos, achieves deep fusion and accurate quantification of multimodal data, and improves the accuracy of sleep state recognition. On the other hand, through an intelligent decision-making module, the multidimensional fused data is transformed into specific meridian targets and intervention scenarios. The system can automatically call and issue personalized hardware control commands. This breakthrough breaks down the data barriers between assessment and intervention, enabling dynamic adaptive configuration of hardware parameters and completely resolving the technical shortcomings of existing intervention equipment parameters being fixed and unable to be dynamically adjusted according to individual differences. Furthermore, through the collaborative intervention module, not only is automated spatial precise positioning of the semi-enclosed cabin and the smart headgear achieved, but also the activation sequence and parameter linkage of multiple energy output devices such as somatosensory music, hydrogen-oxygen, and transcranial magnetic stimulation are strictly controlled through preset collaborative intervention protocols. This forms a closed-loop physical intervention path of "environment creation - system regulation - central control," realizing spatial linkage and temporal collaborative control of multiple physical intervention devices, and significantly improving the efficacy of clinical intervention. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 A schematic diagram of some components of a sleep state recognition and intervention device;

[0048] Figure 2 A schematic diagram of the module structure of a sleep state recognition and intervention system;

[0049] Figure 3 A schematic diagram of the hierarchical architecture of the sleep state recognition and intervention system;

[0050] Figure 4 A schematic diagram of an actual measured electroencephalogram (EEG) curve for a user.

[0051] Figure 5 A schematic diagram of the focus and relaxation index curves for a user in actual testing;

[0052] Figure 6This is a schematic diagram showing some of the measured monitoring data and indicator analysis results for a certain user.

[0053] Figure 7 This is a sample report generated by the system based on user data. Detailed Implementation

[0054] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0055] This application aims to address the shortcomings of existing sleep assessment methods, such as a lack of multimodal data fusion, strong reliance on subjectivity, and limited intervention methods, fixed parameters, cumbersome operation, and lack of coordination. This application intelligently generates personalized treatment prescriptions through comprehensive multimodal data assessment, while integrating multiple physical intervention methods such as hydrogen-oxygen inhalation, transcranial magnetic stimulation, somatosensory music / vibration, and heat therapy, constructing a complete technological closed loop from multi-dimensional assessment, intelligent decision-making, collaborative intervention to efficacy verification.

[0056] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0057] See Figure 1In one exemplary embodiment, this application provides a sleep state recognition and intervention device, the physical entity of which includes a computer device 101, a multi-parameter acquisition terminal (not shown in the figure), a semi-enclosed cabin 102, and various intervention devices (not shown in the figure) built into the cabin. The multi-parameter acquisition terminal includes: a finger clip pulse oximeter (collecting SpO2), an arm blood pressure cuff (collecting SBP / DBP / MBP), a pulse oxygen sensor (simultaneously collecting pulse rate HR and derived HRV), a non-invasive dry electrode EEG sensor (collecting resting-state EEG of the forehead and earlobe), and an interactive touch screen (used to guide the user to complete the questionnaire). The above-mentioned multi-parameter acquisition terminal communicates with the computer device via Bluetooth / WiFi / serial bus. The computer device includes a display, an industrial control computer, and necessary input / output devices. The industrial control computer includes at least a memory, a processor, and a computer program stored in the memory and executable on the processor. The industrial control computer is equipped with the sleep state recognition and intervention system of this application. When the processor executes the computer program, it can realize the functions defined by the multimodal data acquisition module, the intelligent decision-making module, and the collaborative intervention module. The semi-enclosed cabin houses: a reclining mechanism driven by a stepper motor, an intelligent headgear driven by a servo motor (with a built-in transcranial magnetic stimulation coil and bio-light wave emitter), a hydrogen-oxygen intervention device (including an oxygen / hydrogen generator and a nasal cannula), a somatosensory music bone conduction vibrator, an ambient lighting module, and a heat therapy module. Computer equipment controls the execution of the above hardware via relays and motor drive circuits. The core of this application lies in the first-ever integration of Western medicine psychological sleep assessment with traditional Chinese medicine meridian localization through multimodal data fusion, achieving a complete closed loop of "assessment-decision-meridian targeting-intervention." Details are as follows.

[0058] See Figure 2 This application provides a sleep state recognition and intervention system, comprising: a multimodal data acquisition module 201, an intelligent decision-making module 202, and a collaborative intervention module 203. The intelligent decision-making module 202 is connected to both the multimodal data acquisition module 201 and the collaborative intervention module 203. The hierarchical architecture of the sleep state recognition and intervention system is as follows: Figure 3 As shown, it is divided into a data acquisition layer, a data processing and decision-making layer, an intervention and execution layer, and a human-computer interaction layer.

[0059] The multimodal data acquisition module 201 is used to acquire the user's physiological data, electroencephalogram (EEG) signal data, and assessment scale data as multimodal data.

[0060] During the multimodal data acquisition and initial monitoring phase, the user sits in the semi-enclosed cabin of the treatment device, and the system guides them to complete subjective scale assessments (Sleep Psychological Comprehensive Scale, SDS, GAD-7); at the same time, objective physiological data (such as blood pressure, blood oxygen, pulse rate, brain waves and HRV, etc.) in the resting state are collected synchronously through sensors such as EEG and multi-parameter monitors, forming a multi-dimensional raw dataset.

[0061] Before intervention, this application first collects physiological data from patients in a resting state using an independent multi-parameter acquisition terminal. Unlike assessments based on a single indicator, this application constructs a four-dimensional linkage assessment model encompassing blood oxygen, blood pressure, heart rate, and heart rate (HRV). This model aims to provide a quantitative basis for subsequent personalized treatment by analyzing physiological data to understand the patient's sleep structure and psychological stress level. The collected physiological data are shown in Table 1.

[0062]

[0063]

[0064]

[0065] The Psychological Stress Index (PSI) is calculated using PPGA and HBI, and its value ranges from 0 to 100.

[0066] Furthermore, a non-invasive dry electrode sensor was used to collect the user's resting-state EEG signals through contact points on the forehead and earlobe. The system collected five types of brainwaves: delta waves, theta waves, alpha waves, beta waves, and gamma waves, as shown in Table 2. The system focused on monitoring and analyzing four core brainwave frequency bands (alpha, beta, delta, and theta waves) strongly correlated with psychological sleep states, and determined the user's current level of consciousness through waveform energy distribution. A user's measured EEG curve is shown below. Figure 4 As shown, the horizontal axis represents time, and the vertical axis represents brainwave frequency.

[0067]

[0068] The system calculates the energy proportion of each frequency band in real time and defines the dominant wave (DominantWave) – the waveform with the highest energy value – as the dominant state label of the current user; if the dominant wave is a β wave, the user is determined to be in a high wake-up state.

[0069] Traditional EEG frequency is defined as the "number of wave peaks per unit time (Hz)," which is greatly affected by individual differences and signal amplitude (vertical axis), leading to poor assessment consistency across devices and populations. To address this, this application proposes a normalized EEG frequency calculation method to eliminate amplitude interference and improve assessment stability.

[0070] Algorithm definition: ;in This indicates the total number of signal changes per unit time. This indicates the number of times a peak occurs per unit of time; This represents the normalized brainwave frequency. This ratio It reflects the "regularity" of brain electrical activity rather than simply the "speed," and can more accurately map the tension and relaxation state of the autonomic nervous system, without being affected by changes in signal gain.

[0071] Furthermore, in addition to physiological and EEG monitoring, this application integrates a standardized psychological scale assessment function. This function supports manual step-by-step responses (to prevent accidental activation and ensure data authenticity). Through the SDS, GAD-7, and a customized sleep questionnaire, a three-dimensional assessment system of "emotion-psychology-behavior" is constructed to correct for deviations in physiological data and provide a psychological basis for treatment plans. The assessment scale data includes the user's scores on the Zung Self-Rating Depression Scale (SDS), the Generalized Anxiety Scale (GAD-7), and the customized sleep questionnaire (ICSD-3).

[0072] The intelligent decision-making module 202 is used to perform fusion analysis on multimodal data, calculate the comprehensive psychological sleep stress index, and, based on the judgment conclusions of five dimensions—attention state, stress status, sleep problems, depression level, and anxiety level—apply a preset dynamic adjustment coefficient to calculate the score of each meridian and determine the core symptom meridian.

[0073] The intelligent decision-making module performs fusion analysis on multimodal data to calculate a comprehensive psychological sleep stress index, specifically including the following processes 1.1 to 1.9.

[0074] Process 1.1: Using the improved eSense digital measurement algorithm, EEG signal data is converted into a dual-dimensional quantitative index of focus and relaxation, and the EEG imbalance index is further calculated. .

[0075] The intelligent decision-making module of this application integrates an improved eSense digital measurement algorithm, which can convert raw EEG signals into an intuitive dual-dimensional quantitative index (0-100 points) of attention and relaxation. The index definition and evaluation range are shown in Table 3. A user's measured attention and relaxation index curves are shown below. Figure 5 As shown, the horizontal axis represents time, and the vertical axis represents the exponent value.

[0076]

[0077] Furthermore, the brainwave imbalance index The calculation formula is: ;in This indicates a focus on quantification indices; This indicates a relaxation of the quantitative index.

[0078] Furthermore, the system automatically identifies users' pathological tendencies by calculating the gap between their level of focus and relaxation. (Gap Index) The calculation formula is: The value is used to determine anxiety or depression tendencies, ranging from 0 to 100. The determination method is shown in Table 4.

[0079]

[0080] Step 1.2: Calculate the arithmetic sum of the scores for all items in the customized sleep questionnaire as the actual total score of the sleep questionnaire. And further calculate the sleep scale score. .

[0081] The customized sleep questionnaire in this application is designed based on ICSD-3 (International Classification of Sleep Disorders, 3rd Revision), which not only assesses symptoms but also classifies etiologies. This application has designed a proprietary multidimensional sleep questionnaire that includes, but is not limited to, the following dimensions: subjective sleep quality, daytime dysfunction, medication use, sleep sequence, physical symptoms, environment, and cognition.

[0082] Specifically, for the classification logic (qualitative analysis), the system automatically identifies the sleep disorder subtype based on the user's selected "most significant sleep problem":

[0083] Type 1 (Insomnia): Difficulty falling asleep / early awakening / light sleep;

[0084] Type 2 (Sleepiness): Excessive daytime sleepiness / acute sleepiness;

[0085] Type 3 (Rhythmic): Shift work / Jet lag / Irregular sleep;

[0086] Subtype 4 (Respiratory type): Snoring / Sleep apnea (focused on blood oxygenation module);

[0087] Subtype 5 (Motor / Abnormal): Restless legs / sleepwalking / teeth grinding.

[0088] For the quantitative scoring logic (quantitative analysis), the total score is calculated as follows: The system calculates the arithmetic sum of the scores for all items (subjective sleep, daytime function, medication, time, disorder, rhythm, physical condition, environment, cognition) in the customized sleep questionnaire to obtain the actual total score of the sleep questionnaire, which has a maximum score of 228 points. The parameter is represented in the formula. Then, the system will Substituting into the normalization formula, the percentage-based sleep score is calculated and denoted as . : .

[0089] The criteria for determining the severity are as follows:

[0090] ≤25 points: No obvious problems;

[0091] 25< ≤50: Mild sleep problems;

[0092] 50< ≤75: Moderate sleep problems;

[0093] >75: Severe sleep problems.

[0094] The sleep scale score The calculation formula is: .

[0095] Procedure 1.3: Calculate the total score of the Generalized Anxiety Scale. And further calculate anxiety scores. .

[0096] This application uses the Generalized Anxiety Scale (GAD-7) for rapid screening of anxiety symptoms and their severity. The GAD-7 scoring mechanism is as follows.

[0097] Scoring rules: No points at all = 0 points, several days = 1 point, more than half the days = 2 points, almost every day = 3 points.

[0098] Total Score Calculation: The GAD-7 scale Total Score (GAD) is the sum of the scores of the 7 items, ranging from 0 to 21 points. It is represented as a parameter in the formula. .

[0099] The clinical grading criteria for the Anxiety Status Assessment (GAD-7 digital algorithm) are shown in Table 5.

[0100]

[0101] The anxiety score The calculation formula is: .

[0102] Procedure 1.4: Calculate the total score of the Zung Self-Rating Depression Scale. And further converted into SDS standard scores. As a depression score .

[0103] The Zung Self-Rating Depression Scale (SDS) was used to assess the depressive state (SDS digital algorithm), which includes 20 items covering affective, somatic, and psychomotor disorders. The scoring mechanism (including reverse scoring logic) is as follows.

[0104] Positive scoring questions (unmarked): No or very little time = 1 point, a small amount of time = 2 points, a considerable amount of time = 3 points, the vast majority of time = 4 points.

[0105] Reverse scoring questions (with numbers, such as "I think the morning is the best time of day"): Use reverse mapping (4→1, 3→2, 2→3, 1→4) to eliminate response set bias.

[0106] Raw Score Calculation: The Raw Score (SDS) of the SDS scale is the sum of the scores of the 20 items, ranging from 20 to 80 points. It is represented as a parameter in the formula. .

[0107] Standard score conversion (standardization): ;in This represents the "rounding to the nearest integer" function. Multiplying by 1.25 maps the total scale score to a 0-100 point percentage range, facilitating cross-module comparisons. The clinical grading criteria for the SDS StandardScore (SDS) are shown in Table 6, and are denoted as parameters in the formula. .

[0108]

[0109] The depression score The calculation formula is: .

[0110] Procedure 1.5: Based on the collected heart rate and heart rate variability data, a stepwise scoring method is applied according to the stress threshold of the autonomic nervous system to determine the heart rate score. This heart rate score is then mapped to a percentage scale as the heart rate abnormality score. .

[0111] The system assigns a tiered score (out of 4 points) based on the collected resting heart rate mode / mean (HR) and heart rate variability (HRV) data, compared with the stress threshold of the autonomic nervous system. The calculation and scoring logic is as follows.

[0112] HR abnormality criteria: "Heart rate that is too fast (>100 beats / min) or too slow (<60 beats / min) is considered a high-risk scoring item." For example, a heart rate between 60-100 beats / min scores 0 points; a borderline heart rate (e.g., 50-60 or 100-110 beats / min) scores 1-2 points; and a severely abnormal heart rate (<50 or >110 beats / min) scores 3-4 points.

[0113] HRV Assisted Correction: The lower the HRV value, the worse the autonomic nervous system's regulatory capacity. If the HRV is extremely low (sympathetic nervous system is extremely excited), the system will further increase the heart rate abnormality score on top of the HR score.

[0114] Normalization: =Heart rate score / 4 × 100, mapping 0-4 points to a percentage scale of 0-100.

[0115] Procedure 1.6: Based on the magnitude of the decrease in blood oxygen concentration and the risk of apnea, a stepwise scoring system is applied to determine the blood oxygen score, which is then mapped to a percentage scale as the abnormal blood oxygen score. .

[0116] Based on the collected baseline blood oxygen value, average blood oxygen value, and oxygen reduction index (SpO2) data, the system assigns a tiered score (out of 3 points) according to the magnitude of the decrease in blood oxygen concentration and the risk of sleep apnea. The calculation and scoring logic is as follows.

[0117] Normal range: Average blood oxygen >95%, oxygen depletion index is normal, score 0.

[0118] Mild hypoxia / risk: Average blood oxygen saturation is between 90% and 95%, or there is a slight decrease in oxygen, indicating mild sleep structure disruption, scoring 1-2 points.

[0119] Severe hypoxia / high risk: Average blood oxygen <90% (especially OSAHS patients drop to 70%-60%), indicating a serious risk of sleep apnea and microarousal, directly leading to central nervous system hypoxemia, scoring 3 points.

[0120] Normalization: =Blood oxygen score / 3 × 100, mapped to a percentage system of 0-100.

[0121] Step 1.7: Divide blood pressure ranges into systolic, diastolic, and normal, and assign scores accordingly to determine blood pressure scores. Then, map the blood pressure scores to a percentage scale to serve as blood pressure abnormality scores. .

[0122] Based on the collected systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean blood pressure (MBP) data, the system divides blood pressure ranges into high pressure (≥140 / 90), low pressure (<90 / 60), and normal, and assigns a score (out of 3 points) accordingly. The calculation and scoring logic is as follows.

[0123] Normal range: Systolic blood pressure 90-139 and diastolic blood pressure 60-89, score 0.

[0124] Borderline / Mild Abnormal: Blood pressure is at the high end of normal (e.g., 130-139 / 80-89) or slightly below the lower limit of normal, indicating a risk of autonomic nervous system dysfunction, scoring 1-2 points.

[0125] Pathological range: Reaching the high pressure standard (≥140 / 90) or low pressure standard (<90 / 60) indicates "positive correlation with the risk of developing anxiety disorder or schizophrenia in the future" or "leading to insufficient brain perfusion", which is considered high risk and scores 3 points.

[0126] Normalization: = Blood pressure score / 3 × 100, mapped to a percentage system of 0-100.

[0127] Process 1.8: Using Formulas Calculate attention deficit score This is to supplement EEG indicators.

[0128] Process 1.9: The system normalizes the multidimensional heterogeneous data to a percentage scale, and performs weighted summation according to preset weights to calculate the comprehensive psychological sleep stress index. The calculation formula is: .

[0129] After acquiring physiological data, electroencephalogram (EEG) signal data, and assessment scale data, this application uses an intelligent decision-making module to execute a data fusion algorithm to generate a Total Psychological Sleep Stress Index (TPSSI), and outputs a grading conclusion and intervention recommendations based on this index. The TPSSI is expressed in the formula as follows: The higher the score, the greater the psychological sleep stress. The specific grading criteria are shown in Table 7.

[0130]

[0131] The system integrates multi-source data using a weighted average method, with weights allocated based on clinical relevance (30% for sleep behavior, 50% for anxiety and depression, and 20% for physiological factors) to avoid bias from single data sources. For example, if an EEG shows "high anxiety" but the scale shows "no anxiety," the system classifies it as "physiological tension" and prioritizes initiating hydrogen-oxygen therapy instead of rTMS. Some actual monitoring data and indicator analysis results from a user are shown below. Figure 6 As shown, the TPSSI score was 94, indicating that the patient is currently in a harmonious state of mind and body, with coordinated bodily functions, stable emotions, and good sleep quality.

[0132] In addition, this application not only provides a total score TPSSI, but also breaks down the etiology into four dimensions—cognitive dimension, Qi and blood / body dimension, environmental dimension, and rhythm dimension—through a four-dimensional attribution analysis algorithm to guide precise intervention.

[0133] The cognitive dimension score is calculated based on the difference between specific cognitive questions and sleep quality, quantifying the "catastrophic cognition of insomnia." The calculation logic is as follows.

[0134] The system first calculates the reference baseline values: the average of the SDS StandardScore (SDS), the GAD-7 TotalScore (GAD), and the daytime dysfunction score.

[0135] Then, the absolute value of the difference between the reference baseline value and the subjective sleep quality score SleepScore100 is calculated.

[0136] The difference score is determined based on the absolute value of the difference: if the difference is <5, the difference score is 0; if the difference is <10, the difference score is 2; if the difference is <15, the difference score is 4; if the difference is ≥15, the difference score is 6.

[0137] Finally, the cognitive dimension score is calculated using the CognitiveScore (represented as a parameter in the formula). The final percentage score is calculated based on the difference between specific cognitive questions and sleep quality, and is obtained using the following formula: The CognitiveScore (0-100) is used to determine whether CBT-I (Cognitive Behavioral Therapy for Insomnia) content needs to be implanted.

[0138] The Blood & Physical Score, representing the body's Qi and Blood dimensions, is derived from scores on body pain, cold and heat, and bowel movements. Specifically, it's calculated by adding the scores for all "About the Body" questions (total 24 points) and converting them to a percentage scale (0-100). This score is used to determine imbalances in meridians and suggests TCM treatment or medication recommendations.

[0139] The EnvironmentalScore refers to scores for bedding, temperature and humidity, and noise levels. The specific calculation method involves adding up the scores for all "Environment" items (total score 4.5 points) and converting it to a percentage. The EnvironmentalScore is used to generate home environment improvement suggestions.

[0140] The RhythmScore indicates the regularity of one's sleep patterns. It is calculated by adding the scores for each item on sleep rhythm (total score 4.5 points) and converting the result to a percentage. The RhythmScore is used to develop light therapy or sleep pattern resetting plans.

[0141] To accurately guide treatment, the system not only outputs the total score TPSSI, but also generates a four-dimensional attribution profile based on the above algorithm to locate the root cause of the lesion. The specific judgment criteria are shown in Table 8 below.

[0142]

[0143] Based on this, the intelligent decision-making module also uses the judgment conclusions of five dimensions, namely attention state, stress state, sleep problems, depression level and anxiety level, to calculate the scores of each meridian and determine the core symptom meridians by applying preset dynamic adjustment coefficients, specifically including the following processes 2.1 to 2.4.

[0144] Process 2.1: Obtain the judgment conclusions of the five dimensions, and use the medium judgment conclusion as the baseline to apply the adjustment coefficient to perform weighted calculation of the standard scores of the twelve meridians.

[0145] Specifically, the system uses a "moderate judgment conclusion" as the baseline (coefficient of 1.0) and adjusts the scores according to the severity of the five dimensions. The judgment dimensions and adjustment rules are shown in Table 9 below.

[0146]

[0147] Based on the judgment conclusions of five core dimensions, the system compares them with a preset "standard meridian score table" and calculates the weighted score of each meridian by dynamically adjusting coefficients. The adjustment range is defined as follows.

[0148] Apply the ±25% rule: if the conclusion of any dimension is one level higher or lower than the average, the corresponding meridian standard score is multiplied by 1.25 (increase) or 0.75 (decrease).

[0149] Apply the ±50% rule: If the anxiety level dimension is determined to be a special case of no anxiety, the corresponding meridian standard score is multiplied by 0.5 (halved).

[0150] Step 2.2: Add up the scores of the same meridian calculated from the five dimensions and average them to obtain a preliminary comprehensive score.

[0151] The single-dimensional score calculation method is as follows: look up the standard score of each meridian under that dimension in the table, multiply it by the adjustment coefficient of that dimension, and get the meridian score under that dimension.

[0152] Dimensional average: The scores of the same meridian calculated from the five dimensions are added together and divided by 5 to obtain the preliminary comprehensive score of the meridian.

[0153] Process 2.3: Introduce the core meridians corresponding to the psychological diagnosis and treatment classification, and multiply the preliminary comprehensive score by 0.75 to correct the classification.

[0154] The psychological classification correction method is as follows: introduce "psychological diagnosis and treatment classification", systematically retrieve the core meridians corresponding to the classification, and multiply the preliminary comprehensive score of the meridian by 0.75 again (reducing by 25%).

[0155] Step 2.4: Sort the final scores of all meridians and determine the meridian with the lowest score as the core symptom meridian.

[0156] Final ranking: The final scores of all twelve meridians are ranked, and the one with the lowest score is determined as the "core symptom meridian" (i.e. the meridian that most needs intervention).

[0157] Traditional Chinese medicine believes that "the heart governs the mind, the liver governs the free flow of qi, and the spleen governs thought," and that psychological and sleep problems are directly related to the flow of qi and blood in the meridians. This application achieves an isomorphic correspondence between "psychology-physiology-meridians" by mapping multimodal data to the twelve meridians (e.g., anxiety corresponds to 'fire / ministerial fire', and depression corresponds to 'wood'). For example, anxiety (ministerial fire) corresponds to the Sanjiao meridian of the hand (Shaoyang), which requires high-concentration hydrogen and oxygen to regulate qi; depression (wood) corresponds to the liver meridian of the foot (Jueyin), which requires low-frequency rTMS to inhibit the overactive anger center.

[0158] This application, while calculating the Comprehensive Psychological Sleep Stress Index (TPSSI), further combines traditional Chinese medicine theory and uses multi-dimensional data fusion to locate the core symptom meridians most closely related to the user's current psychological state, providing precise coordinates for subsequent physical interventions (such as transcranial magnetic acupoint stimulation and bio-light wave introduction).

[0159] In addition to the comprehensive calculations mentioned above, the system also uses specific questions from the scale to perform point-to-point meridian positioning, which includes the determination of six attributes and meridian targeted positioning.

[0160] The method for determining the six elements is as follows: The system removes 3 neutral questions from the 27 questions in SDS and GAD-7, leaving 24 questions that correspond to the six elements (wood, fire, earth, metal, water, and fire).

[0161] Scoring rules: Each question is worth 1-4 points depending on its difficulty.

[0162] Grouped summation: Add up the scores of the four questions corresponding to each row.

[0163] Judgment principle: The line with the highest score is the user's dominant psychological attribute; if the scores are the same, they are ranked according to the priority of wood > fire > fire > earth > metal > water.

[0164] The meridian-targeting positioning method is as follows: the six elements correspond to two meridians, and each psychological attribute corresponds to two meridians (such as "wood" corresponding to the Foot Shaoyang Gallbladder Meridian and the Foot Jueyin Liver Meridian).

[0165] Question Affiliation: Each meridian corresponds to a specific question number.

[0166] Final decision: Calculate the sum of the scores for the questions corresponding to the two meridians under this psychological attribute, and the meridian with the higher score is the representative meridian for this attribute. If the scores are the same, prioritize the Yang meridian (e.g., the Gallbladder meridian is prioritized over the Liver meridian).

[0167] Based on the input multimodal data, the system of this application executes intelligent decision-making logic, and according to the above calculation results, automatically generates a "Psychological Sleep Evaluation Report Form". As a human-computer interaction interface, this report form is a visual result for doctors and users and contains the following content:

[0168] Comprehensive evaluation and grading: Comprehensive Psychological Sleep Stress Index (TPSSI) and four-dimensional attribution portrait; specifically including core conclusions: total score and grading of TPSSI; and attribution analysis: cognitive, qi and blood, environment, and rhythm radar charts;

[0169] Meridian target location: Automatically determine and output the meridians of the core symptoms (such as "Heart Meridian of Hand少阴") and their manifestations, and clarify the pathological root causes.

[0170] Life conditioning prescription: Output diet, work and rest, psychological adjustment, exercise guidance, and health preservation suggestions corresponding to this meridian.

[0171] Among them, the specific hardware device parameters are not displayed in this report form, but are automatically called by the system background. In a specific application example, a sample report page generated by the system of this application is as Figure 7 shown. This sample report is for reference only, and the actual presentation can be fine-tuned according to reading habits without affecting the logical presentation. For example, based on common sense cognition, the higher the score, the better the overall health. Therefore, the 100 - method is adopted, and finally the score is verified, that is: ① ≤ 25, representing severe psychological sleep stress [severe stress]; ② 25 < X ≤ 50, representing moderate psychological sleep stress [moderate stress]; ③ 50 < X ≤ 75, representing mild psychological sleep stress [mild stress]; ④ > 75, representing no psychological sleep stress [physical and mental balance]. It is only a matter of the score display form and has nothing to do with the original actual score calculation, etc.

[0172] In addition, in order to prevent concealment or false reporting, the intelligent decision-making module is also used to execute the following physiological-psychological cross-validation logic.

[0173] Consistency check: When the evaluation scale data shows "severe insomnia (high SDS score)", but the electroencephalogram signal monitoring shows that "the proportion of δ wave energy and heart rate variability are in the normal range", it is determined as "subjective insomnia", and an intervention scenario containing cognitive behavioral therapy audio is automatically matched, and the treatment focuses on cognitive correction.

[0174] Paradoxical warning: When the evaluation scale data shows "no snoring symptoms (low questionnaire score)", but the blood oxygen saturation monitoring shows that "the nocturnal oxygen desaturation index is higher than the preset threshold (low SpO2)", it is determined as "obstructive sleep apnea", an in-depth sleep scenario is automatically matched, and the flow parameter of the hydrogen-oxygen intervention device is forcibly increased to 3.0 L / min, and at the same time a polysomnogram reexamination prompt is generated.

[0175] Based on TPSSI scores and four-dimensional attribution results (cognition / qi and blood / environment / rhythm), this application system automatically triggers the TCM meridian targeted positioning process, transforming the abstract 'stress index' into specific 'meridian target points', providing precise coordinates for the intervention layer.

[0176] The collaborative intervention module 203 controls devices including a semi-enclosed cabin, a smart headgear, a transcranial magnetic stimulation device, and a hydrogen-oxygen intervention device. It is used to match the target intervention scenario according to the comprehensive psychological sleep stress index and the meridian matching of core symptoms, and to control the semi-enclosed cabin and the smart headgear to perform physical linkage, and to control the transcranial magnetic stimulation device and the hydrogen-oxygen intervention device to perform physical intervention operations according to the preset collaborative intervention protocol.

[0177] This application requires data collection in three dimensions before intervention: first, multi-parameter collection of vital signs via wearable sensing devices; second, extraction of EEG signal features and quantification of mental state via dry electrodes; and third, comprehensive assessment of psychological sleep stress based on standardized scales. Based on the above three-dimensional assessment results, the system automatically matches and outputs the corresponding multimodal physical intervention plan. The specific method for matching and outputting the intervention plan is shown in Table 10 below.

[0178]

[0179] Referring to Table 10, the system automatically matches the corresponding intervention scenario plan based on the determined "core symptom meridians" and "psychological attributes." This application adopts a closed-loop logic of "assessment-decision-hardware linkage-scenario triggering." After the system determines the core symptom meridians, it not only calls the intervention parameters at the software level, but also links the semi-enclosed cabin hardware to automatically adjust the physical environment and execute the preset multi-technology composite plan.

[0180] The collaborative intervention module controls the semi-enclosed cabin and the smart helmet to perform physical linkage, specifically including:

[0181] After the intervention is initiated, the semi-enclosed cabin is controlled to automatically close to form a closed intervention space, isolate external interference, and maintain a stable temperature, humidity and sound field environment.

[0182] The intelligent headgear is lowered non-contactly above the user's head to accurately locate the transcranial magnetic stimulation unit and / or bio-light wave emission unit, ensuring that it acts on the target brain region;

[0183] The reclining mechanism inside the semi-enclosed cabin is adjusted to a 15° slightly tilted flat position to promote blood circulation throughout the body, reduce spinal pressure, and induce the body to enter a relaxed state.

[0184] Simultaneously activate warm lighting and background music (white noise) to provide psychological comfort and enhance immersion.

[0185] The collaborative intervention module is used to create an immersive intervention environment, following a collaborative intervention sequence of "environment creation → physical and mental relaxation → system regulation → central control". Specifically, the preset collaborative intervention protocol defines the following sequence logic:

[0186] First, the cabin's ambient lighting and motion-sensing music equipment are activated, converting the 16-150Hz mid-low frequency signals into physical vibrations that act on the user through bone conduction, creating an immersive environment.

[0187] The hydrogen-oxygen intervention device is activated simultaneously or subsequently to regulate the whole system according to the preset hydrogen concentration and flow parameters of the matched target intervention scenario, and to initiate hydrogen-oxygen therapy to regulate the whole system oxidative stress.

[0188] Finally, the transcranial magnetic stimulation device is used to perform central nervous system regulation according to the preset stimulation frequency, intensity, and brain region target parameters of the matched target intervention scenario, so as to precisely regulate the target brain region.

[0189] The parameters of each device are synchronized and linked according to the preset scenarios, avoiding the blind application of single technologies. The system has 12 built-in standard intervention scenarios, and the technical combination of each scenario is designed based on meridian attributes and pathological mechanisms. Table 11 below shows the concrete parameters and compound logic.

[0190]

[0191]

[0192]

[0193]

[0194] For the two syndrome types (subjective insomnia and masked sleep apnea) identified in the assessment decision, the corresponding intervention plan is directly matched; for other syndrome types, doctors need to customize their own plans.

[0195] Subjective insomnia (high SDS score + normal delta waves): Automatically match "meditation" scenarios and add cognitive correction audio (such as CBT-I content).

[0196] For concealed sleep apnea (no snoring in the questionnaire + low SpO2): the system automatically matches the "deep sleep" scenario, increases the hydrogen and oxygen flow rate to 3.0 L / min, and prompts for a PSG retest.

[0197] In addition, all scenarios include a heat therapy module with uniformly preset parameters: Temperature: constant 45℃; Location: lumbosacral region and back area (Shenshu acupoint and other back acupoint areas); Duration: synchronized with the intervention. Traditional Chinese medicine believes that "cold causes contraction," and heat therapy can dilate local blood vessels, promote drug / energy penetration, and enhance the intervention effect.

[0198] After the system loads the scenario, hardware linkage (lowering of the hood, closure of the chamber) and software parameters (rTMS, hydrogen and oxygen, etc.) are activated synchronously, requiring no manual intervention. During treatment, if needed, the doctor can adjust the parameters, but must manually fine-tune them within the following ranges:

[0199] rTMS intensity: ±10% step (range: 50%-130% MSO);

[0200] Hydrogen and oxygen flow rate: ±0.5L / min increments (range: 0.5-3.0L / min);

[0201] Heat therapy temperature: ±5℃ increments (range: 35-50℃).

[0202] In addition, users can also switch to the expert preset solution library (such as "Dr. Li's Tranquilizing Solution", "Insomnia Intervention Solution", "Children's Solution") with one click, covering the parameters of the current scenario.

[0203] While performing hardware intervention, the system automatically generates and outputs a non-device intervention lifestyle prescription based on the core symptom meridians identified and their corresponding psychological representations (such as "resentment and hatred"). This prescription includes four dimensions: diet, rest, psychological adjustment, and exercise guidance, aiming to support hardware intervention and consolidate the therapeutic effect from the perspective of lifestyle habits.

[0204] For example, for users whose core meridian is the "Hand Shaoyin Heart Meridian" (psychological representation: resentment and hatred), in addition to activating the "meditation and Zen meditation" scenario, the system will also output the following lifestyle prescriptions.

[0205] Dietary recommendations: It is advisable to eat red and bitter foods (such as red dates, red beans, and bitter melon), and avoid spicy foods, strong tea, and coffee; you can cook porridge with jujube seeds and lily bulbs.

[0206] Recommended schedule: Take a 15-30 minute nap at noon (11:00-13:00); stay away from blue light stimulation from mobile phones, computers and other screens for 1 hour before bedtime.

[0207] Psychological adjustment: Practice mindfulness meditation for 10-15 minutes daily; express emotions through static methods such as writing and drawing, and avoid excessive excitement or laughter.

[0208] Exercise guidance: mainly walking, yoga, and tai chi, with a slight increase in heart rate, but avoid excessive sweating; you can practice the "shaking head and wagging tail to clear heart fire" movement of Baduanjin.

[0209] During the collaborative intervention and dynamic adjustment process, the system reads the "core symptom meridians" from the report, automatically matches and loads pre-stored hardware intervention scenario packages (such as "Hand Shaoyin Heart Meridian" corresponding to the "Meditation" scenario). The system controls each module to work automatically according to the "collaborative intervention protocol" (environment creation → system adjustment → central regulation). During treatment, users can manually fine-tune hardware parameters (such as intensity and flow) based on their sensory feedback, or switch to expert-preset plans. Once the process is complete, i.e., the intervention is finished, the system returns to standby mode. If users need to verify the efficacy, they can initiate the multimodal data collection and initial monitoring process again to compare the improvement in TPSSI scores between the two reports.

[0210] Furthermore, the corresponding hardware of this system incorporates multiple layers of safety protection to ensure the safety and reliability of the treatment process. For example, in the event of a hardware failure, such as a sensor disconnection, the headgear not being returned to its original position, or a gas leak, the system will respond automatically as follows: immediately stop all outputs (rTMS / hydrogen-oxygen / light wave); the headgear will automatically rise and the chamber will unlock; the hydrogen-oxygen supply valve will be shut off; the screen will display "Hardware failure, please contact the administrator." When the user initiates an emergency stop, such as by pressing the physical emergency stop button or the "Stop" button on the touchscreen, the system will respond automatically as follows: immediately shut off all energy outputs; the headgear will rise and the chamber will open; the emergency stop time and parameter status will be recorded, and the system will directly enter standby mode. When an abnormal environment is detected, such as an internal temperature >40℃ or humidity >80%, the system will respond automatically as follows: pause the heat therapy; start the internal cooling fan; the screen will display "Ambient temperature too high, please check."

[0211] The core of this application lies in following a linear process from assessment to decision-making to intervention. First, a multimodal data acquisition module obtains the user's physiological and psychological state data, specifically including: collecting objective physiological data through sensors such as EEG and multi-parameter monitors, while simultaneously guiding the user to complete subjective assessments such as a comprehensive sleep and psychological scale. Based on this fused data, the system intelligently generates a comprehensive assessment report that includes multi-dimensional analysis (such as EEG and heart rate variability analysis).

[0212] During the intelligent decision-making phase, the system further incorporates a TCM meridian-based targeted localization mechanism based on the evaluation results. Through a pre-set rule engine or classification model, the system isomorphically maps Western medical psychological sleep data to the TCM Six Elements theory (wood, fire, earth, metal, water, and ministerial fire), automatically determining the user's core symptom meridians (e.g., "Hand Shaoyin Heart Meridian" corresponds to resentment and hatred, and "Foot Jueyin Liver Meridian" corresponds to anger and blame). This localization not only considers the manifestations of sleep disorders but also deeply integrates the mechanisms of action of various interventions with the pathological logic of meridian qi and blood circulation.

[0213] Based on the aforementioned core symptom meridians, the system automatically generates a structured, personalized treatment prescription. This prescription clearly defines the specific parameter combinations for each intervention module, and its generation logic not only considers the manifestations of sleep disorders but also deeply integrates the mechanisms of action of each intervention method. For example, the prescription is synergistically designed to combine "the antioxidant and anti-inflammatory effects of hydrogen-oxygen inhalation to improve the internal environment," "the regulatory effect of transcranial magnetic stimulation on neural circuits in specific brain regions," and "the relaxing effect of somatosensory music on the autonomic nervous system through bone conduction."

[0214] During the intervention execution phase, the system, based on the generated prescription, controls the activation of each module and operates according to a preset "collaborative intervention protocol." This protocol defines the temporal logic and parameter linkage rules of multimodal intervention methods, achieving a leap from the superposition of single methods to intelligent collaboration. Specifically, the system first activates the cabin and meridian-based somatosensory music to create an environment for relaxation and physical and mental well-being. Using somatosensory music therapy, the system converts the 16-150Hz mid-low frequency signals in the music into physical vibrations, which are then directly applied to the body through bone conduction, stimulating deep tissues and reducing muscle tension.

[0215] Simultaneously, the system collaboratively activates the hydrogen-oxygen intervention module, utilizing the selective antioxidant and anti-inflammatory effects of hydrogen to reduce systemic oxidative stress and inflammatory responses, creating a stable internal environment for sleep. Building upon this, the transcranial magnetic stimulation module intervenes to regulate core neurological processes. This synergistic mechanism forms a complete intervention pathway of "systemic regulation (hydrogen-oxygen) + physical and mental relaxation (somatosensory music) + central regulation (transcranial magnetic stimulation)."

[0216] During treatment, the system supports manual customization and adjustment of parameters based on user feedback to optimize the individual treatment experience.

[0217] Those skilled in the art will understand that all or part of the modules in the above-described embodiments can be implemented by hardware related to computer program instructions. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can implement the module functions as described in the above-described system embodiments. Any reference to memory or other media in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0218] It should be noted that the information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0219] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0220] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A sleep state recognition and intervention system, characterized in that, include: The multimodal data acquisition module is used to acquire the user's physiological data, electroencephalogram (EEG) signal data, and assessment scale data as multimodal data. The physiological data includes the user's blood oxygen, blood pressure, heart rate, and heart rate variability data. The EEG signal data includes delta wave, theta wave, alpha wave, beta wave, and gamma wave data. The assessment scale data includes the user's scores on the Zung Self-Rating Depression Scale, the Generalized Anxiety Scale, and a customized sleep questionnaire. The intelligent decision-making module is connected to the multimodal data acquisition module and is used to perform fusion analysis on multimodal data, calculate the comprehensive psychological sleep stress index, and calculate the scores of each meridian based on the judgment conclusions of five dimensions: attention state, stress state, sleep problems, depression level and anxiety level, and determine the core symptom meridians by applying preset dynamic adjustment coefficients. as well as The collaborative intervention module, connected to the intelligent decision-making module, is used to match the target intervention scenario according to the comprehensive psychological sleep stress index and the core symptom meridians, and to control the semi-enclosed cabin and intelligent headgear to perform physical linkage, and to control the transcranial magnetic stimulation device and hydrogen-oxygen intervention device to perform physical intervention operations according to the preset collaborative intervention protocol.

2. The sleep state recognition and intervention system according to claim 1, characterized in that, The intelligent decision-making module fuses and analyzes multimodal data to calculate a comprehensive psychological sleep stress index, specifically including: Using an improved eSense digital measurement algorithm, EEG signal data is converted into a dual-dimensional quantitative index of focus and relaxation, and an EEG imbalance index is further calculated. ; The arithmetic sum of the scores for all items in the customized sleep questionnaire is taken as the actual total score of the sleep questionnaire. And further calculate the sleep scale score. ; Statistical analysis of the total score of the Generalized Anxiety Scale And further calculate anxiety scores. ; Calculate the total score of the Zung Self-Rating Depression Scale. And further converted into SDS standard scores. As a depression score ; Based on the collected heart rate and heart rate variability data, a stepwise scoring system was used to determine the heart rate score by comparing it with the stress threshold of the autonomic nervous system. The heart rate score was then mapped to a percentage scale as the heart rate abnormality score. ; A stepwise scoring system is used to determine the blood oxygen score based on the magnitude of the decrease in blood oxygen concentration and the risk of sleep apnea. This score is then mapped to a percentage scale as the abnormal blood oxygen score. ; Blood pressure zones are divided into systolic, diastolic, and normal ranges, and scores are assigned accordingly to determine blood pressure scores. These scores are then mapped to a percentage scale to serve as blood pressure abnormality scores. ; Using formula Calculate attention deficit score ;in To focus on quantifiable indices; Using formula Calculate the comprehensive psychological sleep stress index .

3. The sleep state recognition and intervention system according to claim 2, characterized in that, The brainwave imbalance index The calculation formula is: ;in To relax the quantification index.

4. The sleep state recognition and intervention system according to claim 2, characterized in that, The sleep scale score The calculation formula is: .

5. The sleep state recognition and intervention system according to claim 2, characterized in that, The anxiety score The calculation formula is: .

6. The sleep state recognition and intervention system according to claim 2, characterized in that, The depression score The calculation formula is: in This represents the "rounding to the nearest integer" function.

7. The sleep state recognition and intervention system according to claim 1, characterized in that, The intelligent decision-making module, based on assessments across five dimensions—attention state, stress level, sleep problems, depression level, and anxiety level—calculates scores for each meridian using preset dynamic adjustment coefficients to identify the core symptom meridians, specifically including: The judgment conclusions of the five dimensions are obtained. Taking the moderate judgment conclusion as the baseline, the standard scores of the twelve meridians are weighted and calculated using adjustment coefficients: Apply the ±25% rule: if the conclusion of any dimension is one level higher or lower than the average, the corresponding meridian standard score is multiplied by 1.25 or 0.

75. Apply the ±50% rule: If the anxiety level dimension is determined to be no anxiety, the corresponding meridian standard score is multiplied by 0.5; The scores for the same meridian calculated from the five dimensions are added together and averaged to obtain a preliminary comprehensive score; We introduce the core meridians corresponding to the psychological diagnosis and treatment classification, and multiply the initial comprehensive score by 0.75 to correct the classification. The final scores of all meridians are sorted, and the meridian with the lowest score is identified as the core symptom meridian.

8. The sleep state recognition and intervention system according to claim 1, characterized in that, The collaborative intervention module controls the semi-enclosed cabin and the smart helmet to perform physical linkage, specifically including: The semi-enclosed cabin is controlled to automatically close to form a closed intervention space; The intelligent headgear is controlled to be lowered non-contactly above the user's head to accurately position the transcranial magnetic stimulation unit and / or bio-light wave emission unit; Adjust the reclining mechanism inside the semi-enclosed cabin to a 15° slightly tilted flat position.

9. The sleep state recognition and intervention system according to claim 8, characterized in that, The timing logic defined by the preset collaborative intervention protocol includes: First, the cabin ambient lighting and motion-sensing music equipment are activated, converting the 16-150Hz mid-low frequency signal into physical vibrations that are transmitted to the user through bone conduction. The hydrogen-oxygen intervention device is activated simultaneously or subsequently to adjust the whole-body system according to the preset hydrogen concentration and flow parameters of the matched target intervention scenario; Finally, the transcranial magnetic stimulation device is used to perform central nervous system modulation according to the preset stimulation frequency, intensity, and brain region target parameters of the matched target intervention scenario.

10. The sleep state recognition and intervention system according to claim 1, characterized in that, The intelligent decision-making module is also used to execute physiological-psychological cross-validation logic: When the assessment scale data shows severe insomnia, but the EEG signal monitoring shows that the proportion of delta wave energy and heart rate variability are within the normal range, it is judged as subjective insomnia, and an intervention scenario containing cognitive behavioral therapy audio is automatically matched. When the assessment scale data shows no snoring symptoms, but the blood oxygen saturation monitoring shows that the nighttime oxygen reduction index is higher than the preset threshold, it is determined to be concealed sleep apnea. The system automatically matches a deep sleep scenario and forcibly increases the flow rate parameter of the hydrogen-oxygen intervention device to 3.0L / min, while generating a polysomnography re-examination prompt.

Images