Sleep quality evaluation method

Abstract

The invention relates to the technical field of sleep analysis, and discloses a sleep quality evaluation method which comprises the following steps: acquiring sleep evaluation; analyzing the weight of each item of sleep evaluation through an AI model, wherein the AI model dynamically adjusts the weight based on training data; and performing comprehensive calculation on the sleep evaluation, and outputting a sleep quality evaluation result. According to the method, the weight of sleep evaluation is dynamically adjusted, an existing unscientific mode of fixed weight and average weight is thoroughly replaced, it is ensured that the weight fits clinical influence and individual difference, and the accuracy of sleep evaluation is improved.

Description

Technical Field

[0001] This application belongs to the field of sleep analysis technology and relates to a method for evaluating sleep quality. Background Technology

[0002] Existing sleep assessment technologies with weighted calculations generally employ methods such as "artificially set fixed weights" or "average weight distribution," which fail to adapt to the differences in the clinical impact of various sleep items and individual user characteristics. For example, assigning the same weight to "abnormal blood oxygen saturation, which may cause acute hypoxia risk" and "turning over pattern, which only slightly affects sleep continuity," or uniformly applying a fixed ratio of "50% subjective evaluation and 50% objective data" to all users, leads to two core problems: Ignoring the hierarchy of clinical impact: From the perspective of clinical logic in sleep medicine, items such as "arrhythmia, such as ventricular fibrillation / atrial fibrillation" and "acute hypoxia, apnea lasting ≥40 seconds" have a much greater negative impact on sleep quality than "occasional turning over" and "mild sleep talking". However, fixed weights cannot reflect this difference. The warning effect of high-risk items is diluted, while the weight of low-impact items is overemphasized, and the evaluation results lack clinical guidance value. Unable to adapt to individual sleep characteristics: There are significant differences in sleep problems and health backgrounds among different users. For example, for users who suffer from restless legs syndrome for a long time, the impact of "restlessness of the limbs" on sleep quality is much higher than that of ordinary users. For users with a history of cardiovascular disease, the monitoring priority of "ST segment changes on electrocardiogram" should be significantly increased. However, the existing methods have not established a correlation mechanism between individual characteristics and weights, resulting in extremely poor evaluation results and difficulty in meeting the needs of personalized sleep management.

[0003] The weighting of existing technologies relies heavily on the subjective experience of researchers, failing to incorporate clinical data and authoritative guidelines in sleep medicine. This leads to a disconnect between the weighting logic and clinical diagnostic standards. For example, without considering the core diagnostic value of "sleep latency > 30 minutes" in insomnia diagnostic guidelines, the weight of "sleep latency" is arbitrarily set to 0.1. Alternatively, without verifying the correlation between "self-assessment items and objective indicators" based on historical case data, the weighting of strongly correlated indicators such as "post-wake fatigue assessment" and "any one or more items in deep sleep duration" becomes unbalanced. Consequently, the evaluation results cannot provide effective reference for doctors, and the supporting role of sleep quality assessment in the initial screening of sleep disorders is lost. Summary of the Invention

[0004] To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended as a general commentary, nor is it intended to identify key / important components or describe the scope of protection of these embodiments, but rather as a prelude to the detailed description that follows.

[0005] To address the problems existing in the related technologies, this disclosure provides a sleep quality evaluation method to solve the technical problems of rigid weight calculation and lack of clinical evidence in the prior art.

[0006] In some embodiments, a sleep quality evaluation method is provided, the method comprising: acquiring a sleep evaluation; analyzing the weights of each item in the sleep evaluation using an AI model, wherein the AI ​​model dynamically adjusts the weights based on training data; performing a comprehensive calculation on the sleep evaluation, and outputting a sleep quality evaluation result.

[0007] Preferably, the sleep evaluation includes a user's self-assessment of sleep complaints regarding the target monitored sleep and continuous sleep monitoring data for the target monitored sleep throughout the night; wherein the sleep monitoring data includes sleep risk monitoring data, sleep state monitoring data, and sleep event monitoring data.

[0008] Preferably, the AI ​​model is a random forest regression model.

[0009] Preferably, the training data includes sleep case data, user continuous monitoring data, and clinical guideline mapping data. The case data includes complete sleep monitoring reports and doctor's diagnostic conclusions. The user continuous monitoring data is user-tracked self-sleep scoring data and sleep monitoring data to establish a user's personal sleep characteristic database. The clinical guideline mapping data is the weight of the influence of each sleep symptom on sleep quality in insomnia diagnosis guidelines.

[0010] Preferably, the feature variables used by the AI ​​model to calculate the weights of the autonomous sleep scoring evaluation include: The correlation coefficients between each individual sleep score item and the corresponding objective indicators in sleep monitoring data; the reference priority of each individual sleep score item in doctors' clinical diagnosis; and historical data on the items of frequent sleep problems for individual users.

[0011] Preferably, the characteristic variables used by the AI ​​model to calculate the weights of sleep monitoring data include: the correlation between each item and the pathological diagnosis of sleep disorders; the frequency of occurrence of abnormal items for individual users; and the clinical impact of the item data.

[0012] Preferably, the items for the autonomous sleep scoring and evaluation are selected from the core clinical pathology diagnostic items of the hospital's sleep medicine department within a preset time period, which are indicators that users can autonomously perceive and are strongly correlated with sleep quality.

[0013] Preferably, the sleep risk monitoring data includes at least one or more of the following: ST segment changes on electrocardiogram, arrhythmia, apnea, respiratory arrest, acute hypoxia, chronic hypoxia, blood oxygen saturation, and risk of altered consciousness.

[0014] Preferably, the sleep state monitoring data includes at least one or more of the following: sleep onset time, sleep duration, number of nighttime awakenings, and deep sleep duration.

[0015] Preferably, the sleep event monitoring data includes at least one or more of the following: frequency of turning over, turning over pattern, restlessness, sleeping posture, snoring, sleep talking, teeth grinding, and sleep anxiety level.

[0016] The sleep quality evaluation method provided in this disclosure can achieve the following technical effects: The weighting of sleep assessments is dynamically adjusted, completely replacing the unscientific methods of fixed weights and average weights. This ensures that the weights are aligned with clinical impact and individual differences, thereby improving the accuracy of sleep assessments.

[0017] The above general description and the description below are exemplary and illustrative only and are not intended to limit this application. Attached Figure Description

[0018] One or more embodiments are illustrated by way of example with reference to the accompanying drawings. These illustrations and drawings do not constitute a limitation on the embodiments. Elements having the same reference numerals in the drawings are shown as similar elements. The drawings are not to be scaled. And wherein: Figure 1 This is a flowchart of a sleep quality evaluation method provided in an embodiment of this disclosure; Figure 2 This is a flowchart of a specific method for evaluating a user's sleep quality provided in this embodiment of the disclosure. Detailed Implementation

[0019] To provide a more detailed understanding of the features and technical content of the embodiments of this disclosure, the implementation of the embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. The accompanying drawings are for illustrative purposes only and are not intended to limit the embodiments of this disclosure. In the following technical description, for ease of explanation, several details are used to provide a full understanding of the disclosed embodiments. However, one or more embodiments may still be implemented without these details. In other cases, well-known structures and systems may be simplified in their depiction to simplify the drawings.

[0020] The following description and accompanying drawings fully illustrate specific embodiments of the invention to enable those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, procedural, and other changes. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the order of operation may vary. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. The scope of embodiments of the invention includes the entire scope of the claims and all available equivalents thereof. In this document, each embodiment may be referred to individually or collectively with the term "invention," which is merely for convenience and is not intended to automatically limit the scope of application to any single invention or inventive concept if more than one invention is disclosed. Relational terms such as "first" and "second" are used herein only to distinguish one entity or operation from another entity or operation, without requiring or implying any actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, or apparatus that includes said element. The various embodiments described herein are presented in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the methods, products, etc., disclosed in the embodiments, since they correspond to the method section disclosed in the embodiments, the descriptions are relatively simple; relevant details can be found in the method section description.

[0021] To address the problems existing in the related technologies, this disclosure provides a sleep quality evaluation method to solve the technical problems of rigid weight calculation and lack of clinical evidence in the prior art.

[0022] Correspondingly, see Figure 1 This is a schematic diagram of a sleep quality evaluation method according to an embodiment of the present disclosure. The method includes: S100, obtain sleep evaluation.

[0023] Sleep evaluation includes the user's self-assessment of sleep complaints regarding the target monitored sleep and continuous sleep monitoring data for the target monitored sleep throughout the night; wherein, the sleep monitoring data includes sleep risk monitoring data, sleep state monitoring data, and sleep event monitoring data.

[0024] Self-assessment of sleep can be achieved through mobile applications, apps, and smart sleep monitoring devices. Users are guided to complete their self-assessment within one hour of waking the following morning, typically after the completion of the target sleep monitoring cycle. This ensures a complete correspondence between subjective evaluation and the objectively monitored sleep cycle, preventing data misalignment. After the user completes the assessment, the system automatically associates and stores each subjective score with the date, timestamp, and corresponding objective monitoring item ID of the target sleep monitoring cycle. This provides a data foundation for subsequent AI model calculations of the subjective-objective correlation weight. The self-assessment items are selected from core clinical pathology diagnostic indicators in the hospital's sleep medicine department that are perceptible to users and strongly correlated with sleep quality within a preset timeframe.

[0025] Sleep monitoring data is collected through professional sleep monitoring equipment, such as polysomnography (PSG) monitors, and medical-grade smart monitoring devices that can be used at home, such as sleep multi-parameter detectors, sleep breathing detectors, sleep mattresses with physiological sensors, and infrared behavior monitoring cameras. This data is collected "continuous and uninterrupted" objective data throughout the user's target sleep cycle and categorized into three dimensions: "sleep risk, sleep state, and sleep events," ensuring that the data covers the entire chain of "health risk - sleep structure - behavioral events."

[0026] Sleep risk monitoring data should include at least one or more of the following: ST segment changes on electrocardiogram, arrhythmia, apnea, respiratory arrest, acute hypoxia, chronic hypoxia, blood oxygen saturation, and risk of altered consciousness.

[0027] Sleep monitoring data includes at least one or more of the following: sleep onset time, sleep duration, number of nighttime awakenings, and deep sleep duration.

[0028] Sleep event monitoring data includes at least one or more of the following: frequency of turning over, turning pattern, restlessness, sleeping posture, snoring, sleep talking, teeth grinding, and sleep anxiety level.

[0029] S200 acquires sleep monitoring data for the target sleep and scores it according to the severity.

[0030] This step is the core innovation of the present invention. It adopts a random forest regression model, corresponding to claim 3, as the core of weight calculation. Based on three major categories of training data, namely "sleep case data", "user continuous monitoring data" and "clinical guideline mapping data", corresponding to claim 4, it dynamically adjusts the weights of "self-sleep scoring items" and "sleep monitoring data items", completely replacing the unscientific methods of existing "fixed weights" and "average weights", and ensuring that the weights are in line with "clinical influence" and "individual differences".

[0031] The random forest regression model is adopted. This model has the advantages of anti-overfitting, handling high-dimensional features, and outputting feature importance by constructing multiple decision trees and combining their prediction results. It can accurately calculate the weights of different sleep evaluation items, and has strong interpretability, making it easy to trace the weight calculation logic.

[0032] The specific composition of the training data corresponds to claim 4: Sleep case data: Desensitized historical cases from medical institutions and sleep centers, with no fewer than 5,000 cases. Each case includes "complete sleep monitoring data" covering three dimensions: sleep risk, sleep state, and sleep events, "doctor's diagnosis conclusion" such as insomnia or obstructive sleep apnea syndrome, and "doctor's importance rating of each evaluation item" on a scale of 1 to 10. The higher the rating, the greater the impact on the diagnosis. For example, the doctor may rate "ventricular fibrillation / atrial fibrillation" at 10 points and "occasional turning over" at 2 points, thus establishing a basic association between "item-clinical importance" for the AI ​​model. Continuous user monitoring data: At least 30 consecutive nights of sleep monitoring were conducted on target users, collecting daily "self-reported sleep scores" and "sleep monitoring data" to establish a user-specific sleep characteristic database. For example, if a user consistently had "deep sleep percentage < 10%" and "post-wake fatigue rating ≤ 4 points" for 10 consecutive days, then the strong correlation between "deep sleep duration" and "post-wake fatigue" was recorded. If a user experienced "restlessness ≥ 10 times / hour" for 7 consecutive days but had a "wakefulness difficulty rating ≥ 8 points," then the actual impact of "restlessness" on the user was recorded as low, providing data support for individual weight adjustments. Clinical guideline mapping data: Referencing authoritative clinical guidelines such as the "Guidelines for the Diagnosis and Treatment of Insomnia in Chinese Adults" and the "Guidelines for the Diagnosis and Treatment of Sleep Apnea and Hypopnea Syndrome," we extracted the "priority of sleep symptoms" explicitly stated in these guidelines. For example, the guidelines indicate that "apnea lasting ≥40 seconds" and "sleep latency >30 minutes" are core indicators for diagnosing sleep disorders, while "frequency of turning over" is a secondary indicator. These priorities are converted into "weighted benchmark values," such as a benchmark weight of 1.2 for core indicators and a benchmark weight of 0.6 for secondary indicators, ensuring that the AI ​​weights do not deviate from clinical logic.

[0033] Feature variables for calculating the weights of the self-sleep scoring items: Subjective-Objective Correlation Coefficient: Calculate the Pearson correlation coefficient between each self-rated item and its corresponding objective monitoring item. For example, the correlation between "post-wake fatigue assessment" and "percentage of deep sleep duration," and the correlation between "ease of falling asleep assessment" and "sleep onset duration." The closer the absolute value of the correlation coefficient is to 1, the more accurately the subjective item reflects the objective state, and the higher its weight. For example, an item with a correlation of 0.8 has a higher weight than an item with a correlation of 0.3. Clinical diagnostic reference priority: Based on sleep case data, the "citation frequency" of each main scoring item in the doctor's diagnostic report is calculated. For example, the "ease of waking up" evaluation has a citation frequency of 85% in the diagnosis of chronic insomnia, and the "ease of dreaming" evaluation has a citation frequency of 60%. The higher the citation frequency, the greater the value of the subjective item to the clinical judgment and the higher the weight. User-specific frequent data: Based on continuous monitoring data of users, the "abnormal score frequency" of each user's main scoring item is calculated. A score ≤4 is considered abnormal. For example, if a user's "post-wake fatigue rating" is ≤3 for 7 consecutive days, the symptoms are severe, and the weight of this item will be automatically increased by 20%-30%. If a user's "dream intensity rating" is ≥8 for 14 consecutive days, there is no abnormality, and the weight of this item will be reduced to 80% of the basic weight to ensure that the subjective weight is consistent with the user's recent sleep problems.

[0034] Feature variables for calculating the weights of sleep monitoring data items: The correlation between items and pathological diagnoses of sleep disorders: Based on sleep case data, the "diagnostic contribution" of each objective item to a specific sleep disorder is calculated. For example, the correlation between "ventricular fibrillation / atrial fibrillation" and the risk of cardiogenic sleep is 95%, and the correlation between "occasional sleep talking" and sleep disorders is 10%. The higher the correlation, the higher the base weight. For example, the base weight of an item with a correlation of 95% is 1.5, and the base weight of an item with a correlation of 10% is 0.5. Frequency of abnormal items for individual users: Based on continuous monitoring data of users, the "number of times the abnormal level occurs" of each objective item is counted. For example, if a user has "blood oxygen saturation ≤85%" for 3 consecutive days, which is an abnormal level, the weight of this item will be automatically increased by 25%; if a user's "turning frequency" is in the normal level for 21 consecutive days, the weight of this item will be reduced to 90% of the basic weight to ensure that the objective weight is adapted to individual differences. Clinical impact of item data: Based on clinical case data and guidelines, the severity of health consequences corresponding to different levels of each item is analyzed. For example, "apnea ≥ 40 seconds" may lead to acute hypoxic coma, and its clinical impact is 1.4 times that of "apnea 15 seconds". Therefore, the AI ​​model assigns the former a weight of 1.4 times that of the latter. "No sleep all night" has a 5 times greater impact on bodily functions than "normal sleep duration". Therefore, the former has a 5 times greater weight than the latter, ensuring that the weights reflect the "difference in impact".

[0035] S300 performs comprehensive calculations on sleep evaluations and outputs sleep quality evaluation results.

[0036] Based on the sleep evaluation data from step S100 and the AI ​​dynamic weights from step S200, the system outputs scientific and easy-to-understand sleep quality evaluation results through a process of "item weighted score calculation - subjective and objective comprehensive score calculation - quality level classification".

[0037] In one specific embodiment, "a sleep assessment of a 45-year-old female user who complained of chronic fatigue upon waking and had a history of mild hypertension is used as an example." (See also...) Figure 2 The implementation process of the present invention is described in detail, including: S210, to conduct self-sleep scoring and evaluation.

[0038] The user completed the scoring via the app the following morning, and the results are as follows: Ease of falling asleep rating: 8 points (falling asleep in 10-15 minutes, basically normal); Easy to wake up rating: 6 points (woke up once at night, and fell back asleep 5 minutes after waking up); Severity of dreaming: 7 points (I had one dream occasionally, but it did not affect my sleep); Post-awakening fatigue rating: 3 points (persistent fatigue after waking up, requiring more than 3 hours to recover, symptoms are severe).

[0039] S220 is used for sleep monitoring data collection.

[0040] Data collected throughout the night included ECG, respiration, blood oxygen saturation, body position, posture, and body movement. After preprocessing, the data were categorized as follows: Sleep risk monitoring data: ECG ST segment changes: occasional slight changes (grade 2), preliminary score 15 points; arrhythmia: no abnormalities (grade 1), preliminary score 20 points; blood oxygen saturation: 92% (grade 3, mild hypoxia), preliminary score 10 points; other risk items (apnea, altered consciousness, etc.) are all grade 1, preliminary scores 16-20 points; Sleep monitoring data: Sleep onset time: 12 minutes (Level 3), preliminary score 6; Sleep duration: 6.8 hours (Level 2), preliminary score 6; Number of times you wake up at night: 1 time (Level 2), preliminary score 7; Deep sleep duration: 7% (Level 3), preliminary score 2; Sleep event monitoring data: Turning over frequency: 3 times per hour (level 2), preliminary score 6 points; Restlessness: 2 times per hour (level 2), preliminary score 6 points; Other event items are all level 1-2, preliminary score 3-8 points.

[0041] S230 performs dynamic weight calculations for the AI ​​model.

[0042] The training data input phase begins.

[0043] Sleep case data: 5,000 cases including "sleep monitoring of hypertensive patients" were entered. Among them, 72% of the cases were associated with "deep sleep percentage <10%" and "post-wake fatigue". Doctors rated the importance of "deep sleep duration" as 8 points. User continuous monitoring data: Input the user's data for the past 30 days. The number of days with "post-wake fatigue rating ≤ 4 points" reached 24 days (frequent abnormality), and the number of days with "deep sleep percentage < 10%" reached 18 days. Clinical guideline data: Inputting data from the "Guidelines for Sleep Health Management of Hypertensive Patients" clarifies that "insufficient deep sleep" and "nocturnal hypoxia" are the core indicators for sleep assessment of hypertensive patients, and their weight should be higher than that of ordinary users.

[0044] The stage of calculating feature variables and outputting weights.

[0045] Weighting of self-assessment sleep scoring items: "Post-wake fatigue assessment": Correlation coefficient with "deep sleep duration" 0.78, clinical citation frequency 82%, frequent abnormalities → weight 1.4; "Ease of falling asleep assessment": Correlation coefficient with "sleep onset time" 0.65, no frequent abnormalities → weight 0.9; other subjective items weight 0.7-0.8; Weighting of sleep monitoring data items: "Deep sleep duration (Grade 3)": 75% correlation with "post-wake fatigue", frequent abnormalities observed by users → weight 1.3; "Blood oxygen saturation (Grade 3)": 68% correlation with hypertension-related sleep risk → weight 1.2; "ECG ST segment changes (Grade 2)": 30% correlation, no frequent abnormalities observed → weight 0.8; other low-impact items weighted 0.6-0.7.

[0046] S240, comprehensive calculation and result output.

[0047] Weighted score of autonomous sleep assessment (S1): S1 = [(8×0.9) + (6×0.8) + (7×0.7) + (3×1.4)] / [(1.4×10) + (0.9×10) + (0.8×10) + (0.7×10)] × 100 = [7.2 + 4.8 + 4.9 + 4.2] / [14 + 9 + 8 + 7] × 100 = 21.1 / 38 × 100 ≈ 55.5 points; Weighted score of sleep monitoring data (S2): F = (15×0.8 + 20×0.8 + 10×1.2) + (6×0.9 + 6×1.1 + 7×0.9 + 2×1.3) + (6×0.7 + 6×0.8) = 40 + 18.1 + 9 = 67.1 points; S2 = (67.1 / 180)×100≈37.3 points; Based on the overall score and grade, S = 55.5 × 0.4 + 37.3 × 0.6 = 22.2 + 22.38 = 44.58 points → Sleep quality grade "poor" (< 60 points).

[0048] The foregoing description and accompanying drawings fully illustrate embodiments of the present disclosure to enable those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, procedural, and other changes. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the order of operations may vary. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. Moreover, the terminology used in this application is for describing embodiments only and is not intended to limit the claims. As used in the description of embodiments and claims, unless the context clearly indicates otherwise, the singular forms “a,” “a,” “an,” and “the” are intended to equally include the plural forms. Similarly, the term “and / or” as used in this application means including one or more of the associated listed items and all possible combinations thereof. Additionally, when used in this application, the terms “comprising,” “comprise,” and variations thereof, “comprises,” “comprising,” etc., refer to the presence of stated features, integrals, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof. Unless otherwise specified, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, or apparatus that includes said element. In this document, each embodiment may focus on its differences from other embodiments, and similar or identical parts between embodiments can be referred to mutually. For methods, products, etc., disclosed in the embodiments, if they correspond to the method section disclosed in the embodiments, then the relevant parts can be referred to the description of the method section.

[0049] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments of this disclosure. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, units, and processes described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0050] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

Claims

1. A method for evaluating sleep quality, characterized in that, The method includes: Obtain sleep assessment; The weights of each item in the sleep evaluation are analyzed using an AI model, which dynamically adjusts the weights based on training data. The sleep evaluation is comprehensively calculated to output the sleep quality evaluation result.

2. The method according to claim 1, characterized in that, Sleep evaluation includes the user's self-assessment of sleep complaints regarding the target monitored sleep and continuous sleep monitoring data for the target monitored sleep throughout the night; wherein, the sleep monitoring data includes sleep risk monitoring data, sleep state monitoring data, and sleep event monitoring data.

3. The method according to claim 1, characterized in that, The AI ​​model is a random forest regression model.

4. The method according to claim 3, characterized in that, The training data includes sleep case data, user continuous monitoring data, and clinical guideline mapping data. The case data includes complete sleep monitoring reports and doctor's diagnostic conclusions. The user continuous monitoring data is user-tracked self-sleep scoring data and sleep monitoring data to establish a user's personal sleep characteristic database. The clinical guideline mapping data is the weight of the influence of each sleep symptom on sleep quality in insomnia diagnosis guidelines.

5. The method according to claim 4, characterized in that, The feature variables used by the AI ​​model to calculate the weight of the autonomous sleep scoring evaluation include: The correlation coefficient between each individual sleep score item and the corresponding objective indicator in the sleep monitoring data; The reference priority of each individual sleep scoring item in doctors' clinical diagnosis; Historical data on projects involving frequent sleep problems among individual users.

6. The method according to claim 4, characterized in that, The feature variables used by the AI ​​model to calculate the weights of sleep monitoring data include: The correlation between each item and the pathological diagnosis of sleep disorders; Frequency of occurrence of individual user anomaly entries; The clinical impact of the item data.

7. The method according to claim 2, characterized in that, The self-sleep scoring and evaluation items are selected from the core clinical pathology diagnostic items of the hospital's sleep medicine department within a preset time period, and are indicators that users can perceive independently and that are strongly correlated with sleep quality.

8. The method according to claim 2, characterized in that, Sleep risk monitoring data should include at least one or more of the following: ST segment changes on electrocardiogram, arrhythmia, apnea, respiratory arrest, acute hypoxia, chronic hypoxia, blood oxygen saturation, and risk of altered consciousness.

9. The method according to claim 2, characterized in that, Sleep monitoring data includes at least one or more of the following: sleep onset time, sleep duration, number of nighttime awakenings, and deep sleep duration.

10. The method according to claim 2, characterized in that, Sleep event monitoring data includes at least one or more of the following: frequency of turning over, turning pattern, restlessness, sleeping posture, snoring, sleep talking, teeth grinding, and sleep anxiety level.

Images