Data processing method and system based on insomnia behavior

Abstract

The invention discloses a data processing method and system based on insomnia behaviors, and relates to the technical field of health monitoring, and the method comprises the steps: collecting sleep behavior data of a user through a smart bracelet and a sleep diary application; performing exception labeling on the sleep behavior data according to a preset normal range; evaluating an insomnia risk level according to the number of the abnormal labels; corresponding processing flows are triggered according to different risk levels. The system comprises an acquisition module, an evaluation module and a processing module. According to the invention, automatic assessment and graded intervention of the insomnia risk are realized, and the accuracy of insomnia monitoring and the pertinence of intervention are improved.

Description

Technical Field

[0001] This invention relates to the field of health monitoring technology, and in particular to a data processing method and system based on insomnia behavior. Background Technology

[0002] With the accelerating pace of social life and increasing work pressure, insomnia is becoming increasingly common. Insomnia is a common sleep disorder that refers to a subjective experience of dissatisfaction with sleep duration or quality despite adequate sleep opportunities and environment, which affects daytime social functioning.

[0003] Currently, Chinese invention patent application number CN114864078A discloses a method, system, electronic device, and computer-readable medium for processing insomnia data. The method includes: acquiring feature tags of insomnia users; determining a target treatment plan that matches the insomnia user from multiple candidate treatment plans in a preset treatment plan library based on the feature tags; and pushing the target treatment plan to the insomnia user.

[0004] The aforementioned technologies are difficult to collect data in real time, have significant subjective biases, and lack dynamic risk assessment, tiered intervention, insomnia risk prediction, and personalized adjustments, which can easily lead to low data accuracy and poor real-time performance. Summary of the Invention

[0005] The technical problem solved by this invention is that existing technologies are difficult to collect data in real time, have large subjective biases, and do not perform dynamic risk assessment, graded intervention, insomnia risk prediction and personalized adjustment, which easily leads to low data accuracy and poor real-time performance.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: First aspect, a data processing method based on insomnia behavior, comprising the following steps: Step S1, collecting sleep behavior data of users according to different instruments, receiving trigger conditions and marking abnormalities according to the sleep behavior data;

[0007] Step S2: Based on the number of marked abnormalities, proceed to the assessment node to obtain the insomnia risk level;

[0008] Step S3: Based on the insomnia risk level, trigger the corresponding processing flow, and output an analysis report after the processing flow is completed.

[0009] As a preferred embodiment of the data processing method based on insomnia behavior described in this invention, step S1 includes the following sub-steps:

[0010] Step S101: Collect sleep behavior data of users according to different instruments. The sleep behavior data includes sleep onset time data, sleep duration data, number of awakenings data, sleep efficiency data, and deep sleep ratio data.

[0011] Step S102: Obtain the sleep onset time data, sleep duration data, number of awakenings data, and deep sleep ratio data through the smart bracelet;

[0012] Step S103: Obtain sleep efficiency data through a sleep diary application.

[0013] As a preferred embodiment of the data processing method based on insomnia behavior described in this invention, step S1 further includes the following sub-steps:

[0014] Step S104: Based on the sleep behavior data, a normal sleep range is preset; based on the normal sleep range, trigger conditions are received; the trigger conditions are identified; and identification results are obtained. The normal sleep range includes the sleep onset time range, sleep duration range, number of awakenings, sleep efficiency value, and deep sleep ratio value.

[0015] Step S105, and record the data information corresponding to the recognition result according to the data type of sleep behavior data. The triggering conditions include a first triggering condition, a second triggering condition, a third triggering condition, a fourth triggering condition, and a fifth triggering condition.

[0016] Step S106: The first trigger condition is sleep onset time data, the second trigger condition is sleep duration data, the third trigger condition is wake-up frequency data, the fourth trigger condition is sleep efficiency data, and the fifth trigger condition is deep sleep ratio data.

[0017] As a preferred embodiment of the data processing method based on insomnia behavior described in this invention, the anomaly labeling specifically includes:

[0018] In response to the first trigger condition result, the sleep time data is compared with the sleep time interval. If the sleep time data is not within the sleep time interval, it is marked as abnormal.

[0019] In response to the second trigger condition result, the sleep duration data is compared with the sleep duration range. When the sleep duration data is less than or greater than the sleep duration range, an anomaly is marked.

[0020] In response to the third trigger condition result, the number of times the user wakes up is compared with the number of times the user wakes up. If the number of times the user wakes up is greater than the number of times the user wakes up, an anomaly is marked.

[0021] In response to the fourth trigger condition result, the sleep efficiency data is compared with the sleep efficiency value. If the sleep efficiency data is less than the sleep efficiency value, an anomaly is marked.

[0022] In response to the fifth trigger condition result, the deep sleep ratio data is compared with the deep sleep ratio value. When the deep sleep ratio data is less than or greater than the deep sleep ratio value, an anomaly is marked.

[0023] As a preferred embodiment of the data processing method based on insomnia behavior described in this invention, step S2 specifically includes:

[0024] Based on the anomalies, an assessment node is established to obtain the insomnia risk level. The assessment node involves obtaining the number of anomalies based on the anomalies, and then assessing the number of anomalies against a preset insomnia level to obtain the insomnia risk level. The preset insomnia level includes low risk, medium risk, and high risk.

[0025] The low risk is defined as the number being less than or equal to a first threshold.

[0026] The medium risk is defined as the number of items being greater than a first threshold and less than or equal to a second threshold;

[0027] The high risk is defined as the number of [number] items being greater than the second threshold.

[0028] As a preferred embodiment of the data processing method based on insomnia behavior described in this invention, step S3 includes the following sub-steps:

[0029] Step S301: When the insomnia risk level is low, proceed with the routine monitoring process, including adjusting the normal sleep range and updating the insomnia risk level.

[0030] Step S302: When the insomnia risk level is medium risk, the early warning intervention process is initiated, including sending early warning information to the user and providing improvement suggestions.

[0031] Step S303: When the insomnia risk level is high, the expert consultation process is initiated, including automatically matching relevant experts and pushing assessment reports.

[0032] As a preferred embodiment of the data processing method based on insomnia behavior described in this invention, the conventional monitoring process specifically includes:

[0033] Collect basic information about the user, including name, age, gender, and medical history;

[0034] Based on the basic information and sleep behavior data, a personalized monitoring plan is formulated. The personalized monitoring plan includes identifying abnormal sleep behavior data based on the sleep behavior data, obtaining historical sleep behavior data corresponding to the abnormal sleep behavior data, and adjusting the normal sleep range based on the historical sleep behavior data. The historical sleep behavior data is the sleep behavior data of the 30 days prior to the current timestamp of the acquired sleep behavior data.

[0035] As a preferred embodiment of the data processing method based on insomnia behavior described in this invention, the early warning intervention process specifically includes:

[0036] A correlation is established based on pre-stored high-risk sleep behavior data and time. By using the historical sleep behavior data, sleep behavior data and the correlation, the risk of future insomnia is predicted, and the prediction results are obtained.

[0037] If the predicted result is greater than the sixth threshold, the insomnia risk level is increased and an early warning message is issued;

[0038] Provide users with targeted improvement suggestions through a human-computer interaction interface.

[0039] As a preferred embodiment of the data processing method based on insomnia behavior described in this invention, the expert consultation process specifically includes:

[0040] Based on the marked anomalies, corresponding sleep behavior data is obtained. Based on the sleep behavior data, the insomnia risk type is obtained. Based on the insomnia risk type, a corresponding expert is matched, and an analysis report of key sleep indicators is generated. The insomnia risk types include difficulty falling asleep, difficulty maintaining sleep, early awakening, low sleep efficiency, and insufficient deep sleep. The analysis report includes basic information, historical sleep behavior data, improvement suggestions, and doctor information.

[0041] Secondly, a data processing system based on insomnia behavior includes a data collection module, an assessment module, and a processing module;

[0042] The acquisition module is used to collect users' sleep behavior data according to different instruments, and to receive trigger conditions and mark abnormalities based on the sleep behavior data;

[0043] The assessment module is used to enter the assessment node based on the number of marked abnormalities and obtain the insomnia risk level;

[0044] The processing module is used to trigger a corresponding processing flow based on the insomnia risk level, and output an analysis report after the processing flow is completed.

[0045] The beneficial effects of this invention are as follows: Through multi-source data collection and automated analysis, accurate assessment of insomnia risk is achieved. The system collects multi-dimensional sleep behavior data, including sleep onset time, sleep duration, number of awakenings, sleep efficiency, and deep sleep ratio, through smart bracelets and sleep diary applications, establishing a comprehensive sleep quality assessment system that makes the assessment method more accurate and reliable. The system presets normal ranges for various sleep parameters and can quickly identify abnormal sleep patterns through automatic comparison and data annotation, providing a reliable data foundation for subsequent risk assessment. A scientific graded assessment and intervention mechanism is established, dividing insomnia risk into three levels: low risk, medium risk, and high risk, based on the number of abnormal annotations. Differentiated processing procedures are adopted for different levels. This graded processing method avoids over-intervention while ensuring that high-risk users receive timely attention. For low-risk users, the system adopts conventional monitoring methods, achieving dynamic monitoring through continuous data collection and threshold adjustment. For medium-risk users, the system initiates an early warning intervention process, combining historical data to predict risk, issuing early warnings, and providing personalized improvement suggestions. For high-risk users, the system automatically matches them with professional doctors, generates detailed analysis reports, and enables rapid professional intervention. By collecting users' basic information and historical sleep data, the system develops personalized monitoring plans, dynamically adjusting various parameters within their normal ranges based on the user's historical sleep patterns, making the assessment standards more aligned with the user's actual situation. In the early warning intervention phase, the system provides targeted improvement suggestions based on the user's specific sleep problem type. By analyzing the changing trends of historical sleep data, it can predict future insomnia risks and provide early warnings. When the prediction results exceed a threshold, the system automatically raises the risk level and issues an early warning to prevent further deterioration of sleep problems. The generated analysis report includes comprehensive sleep indicators and professional improvement suggestions, making it convenient for users to understand their sleep status and providing important references for doctors' diagnoses. Attached Figure Description

[0046] Figure 1 A flowchart illustrating the steps of a data processing method based on insomnia behavior according to an embodiment of the present invention;

[0047] Figure 2 This is a basic flowchart of a data processing system based on insomnia behavior provided in one embodiment of the present invention. Detailed Implementation

[0048] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0049] This invention provides a data processing method based on insomnia behavior. The method can be applied to devices including but not limited to health monitoring systems and sleep management platforms, and can be used in scenarios including but not limited to personal sleep quality assessment, insomnia risk warning and personalized intervention. It is applicable to scenarios such as family health management and sleep clinics in medical institutions.

[0050] In response to the technical problems of subjective assessment, delayed intervention measures, and lack of systematic approach in traditional insomnia technologies, the inventors propose a data processing method based on insomnia behavior in this invention. The core idea of ​​this invention is to collaboratively collect multi-source sleep behavior data, mark anomalies based on a preset normal range, assess the insomnia risk level based on the number of anomalies, and trigger a graded processing flow to achieve accurate identification and adaptive intervention of insomnia risk, thereby improving the efficiency and effectiveness of sleep health management.

[0051] Example 1, referring to Figure 1 , Figure 1 This is a flowchart illustrating a data processing method based on insomnia behavior provided in an embodiment of the present invention. Figure 1 As shown, in this embodiment, the method includes, but is not limited to, the following steps S1-S3.

[0052] In one embodiment, S1, the user's sleep behavior data is collected from different instruments, and based on the sleep behavior data, trigger conditions are received and abnormalities are marked.

[0053] Explained, this step corresponds to step S1 of claim 1 and steps S101-S103 of claim 2. Sleep behavior data is the foundation for assessing a user's sleep quality. This data is collected through both objective and subjective methods to ensure comprehensiveness and accuracy. Specifically, sleep behavior data includes sleep onset time data, sleep duration data, number of awakenings data, sleep efficiency data, and deep sleep ratio data. These parameters together constitute a quantitative indicator of sleep quality. Sleep onset time data refers to the time when the user actually enters a sleep state, usually expressed in 24-hour format, reflecting the user's sleep-wake cycle. Sleep duration data refers to the total time from falling asleep to waking up, expressed in hours or minutes, measuring sleep sufficiency. Number of awakenings data refers to the number of times the user wakes up during sleep, reflecting sleep continuity. Sleep efficiency data refers to the ratio of actual sleep time to time spent in bed, expressed as a percentage, assessing sleep efficiency. Deep sleep ratio data refers to the proportion of deep sleep stage to total sleep time, expressed as a percentage. Deep sleep is a key stage of restorative sleep. The multi-dimensional collection of these data avoids the limitations of a single indicator and provides a rich information foundation for subsequent analysis.

[0054] Step S101: Collect sleep behavior data of users according to different instruments. Sleep behavior data includes sleep onset time data, sleep duration data, number of awakenings data, sleep efficiency data, and deep sleep ratio data.

[0055] Explained, this approach leverages the collaborative collection of sleep behavior data via smart bracelets and sleep diary apps to ensure a balance between data objectivity and subjective experience. The smart bracelet, a wearable device, incorporates accelerometers and heart rate sensors to continuously monitor the user's physiological parameters and activity levels. The sleep diary app, installed on the user's smartphone, collects sleep-related data through user input. This multi-instrument data collection method utilizes sensor technology and human-computer interaction principles. The smart bracelet provides objective physiological data, while the sleep diary app supplements subjective sleep experience data, thus reducing bias from a single data source. For example, in a typical application scenario, a user wears a smart bracelet to sleep. The bracelet detects body activity levels using an accelerometer. When the activity level remains below a threshold, such as no significant movement within 5 minutes, the system determines that the user has fallen asleep and records the sleep onset time. Sleep duration is calculated by determining the time difference between the sleep onset time and the final wake-up time. The final wake-up time is determined by detecting continuous physical activity, such as the user getting out of bed. The number of wake-ups is determined by monitoring peak limb activity during sleep; if the activity level briefly exceeds a threshold, such as when the user turns over or gets out of bed, the system records a wake-up. The deep sleep ratio is determined by monitoring heart rate variability using a heart rate sensor and using a pre-trained machine learning model, such as a support vector machine or neural network, to analyze the heart rate variability data and estimate the deep sleep ratio. The machine learning model maps heart rate variability features to sleep stages and outputs the deep sleep ratio. The sleep diary application guides users to manually input sleep information each morning through a graphical interface, including bedtime, actual sleep onset time, final wake-up time, and number of nighttime wake-ups. The application automatically calculates the time spent in bed (from bedtime to final wake-up time) and the actual sleep time (from sleep onset time to final wake-up time), subtracts the wake-up time, and then calculates sleep efficiency according to the formula: Sleep Efficiency = (Actual sleep time / time in bed) × 100%. In practice, the smart bracelet typically collects data once per minute to ensure the real-time and continuous nature of the data. The data is transmitted to the central processing system via Bluetooth or wireless network. The data of the sleep diary application is stored locally or in a cloud database for the system to access. This multi-source collection method is based on the principle of data fusion and improves the reliability of the overall data through complementary data sources.

[0056] Step S102: Obtain sleep onset time data, sleep duration data, number of awakenings data, and deep sleep ratio data through the smart bracelet.

[0057] Explained, the smart bracelet's implementation is based on sensor data analysis and pattern recognition principles. Sleep onset time data is obtained by detecting the user's physical activity level using an accelerometer. When the activity level remains below a threshold, the system determines the user has fallen asleep and records this time. The algorithm identifies sleep onset time based on activity patterns, avoiding human error. Sleep duration data is obtained by calculating the time difference between sleep onset and final wake-up time. The final wake-up time is determined by detecting continuous physical activity. Wake-up frequency data is obtained by monitoring peak limb activity during sleep. If the activity level briefly exceeds a threshold, the system records a wake-up. The number of wake-ups is calculated by counting these events. Deep sleep ratio data is obtained by monitoring heart rate variability using a heart rate sensor. The system uses pre-trained machine learning models to analyze variability data and estimate the proportion of deep sleep. For example, the machine learning model maps heart rate variability features, such as low-frequency oscillations, to sleep stages and outputs the proportion of deep sleep. In practice, the sensor data of the smart bracelet is filtered and preprocessed to reduce noise interference. For example, a low-pass filter is used to smooth acceleration data, or frequency domain analysis is used to process variability data. After data acquisition, the data is initially processed by an embedded system, such as a microcontroller, and then transmitted to a central server. For example, during a user's sleep, the smart bracelet may record that the time of falling asleep is 23:30, the sleep duration is 7 hours, the number of awakenings is 2, and the proportion of deep sleep is 20%. These data are used as raw inputs for subsequent use.

[0058] Step S103: Obtain sleep efficiency data through a sleep diary application.

[0059] Explained, the specific implementation of a sleep diary app is based on user input and automatic calculation principles. Every morning, users input their sleep information through the app interface, including bedtime, actual sleep onset time, final wake-up time, and number of nighttime awakenings. The app automatically calculates the time spent in bed and the actual sleep time, then calculates sleep efficiency according to a formula. For example, if a user inputs a bedtime of 22:00, a sleep onset time of 22:30, and a final wake-up time of 6:30, then the actual sleep time is 8 hours, the time spent in bed is 8.5 hours, and the sleep efficiency is (8 / 8.5) × 100% = 94.1% of applications guide user input through graphical interfaces, such as sliders or numeric input boxes, reducing input errors. They also employ data validation mechanisms, such as range checks, to ensure input validity. Data is stored locally or in a cloud database and synchronized with a central system via API. The principle behind this approach is to combine user subjective recollections to compensate for the shortcomings of smart bracelets in calculating sleep efficiency. This is because bracelets may not accurately distinguish between bedtime and actual sleep time. For example, if a user falls asleep after reading in bed, the bracelet may overestimate the actual sleep time. Sleep diary applications, on the other hand, provide more accurate efficiency data through user input.

[0060] Step S104: Based on sleep behavior data, preset a normal sleep range, receive trigger conditions based on the normal sleep range, identify the trigger conditions, and obtain the identification results. The normal sleep range includes the sleep onset time range, sleep duration range, number of awakenings, sleep efficiency value, and deep sleep ratio value.

[0061] Interpretatively, the preset normal sleep ranges are reference values ​​based on medical research and clinical practice, used to determine whether sleep behavior data is abnormal. These ranges can be adjusted according to user groups, such as age and gender, but the following example values ​​are typically used: sleep onset time range of 22:00 to 24:00, which aligns with the human body's biological clock and helps maintain a healthy sleep-wake cycle; sleep duration range of 6 to 9 hours, covering the recommended sleep duration for adults; number of awakenings of no more than 2, as frequent awakenings can affect sleep depth and continuity; sleep efficiency of no less than 85%, as a sleep efficiency below 85% indicates poor sleep quality; and the proportion of deep sleep of 20% to 25%, ensuring effective sleep recovery. The triggering conditions are rules defined internally by the system for data detection. Whether it exceeds the normal range, the system receives these conditions and performs a comparison operation on each sleep behavior data. The identification result includes "normal" or "abnormal", as well as the degree of deviation. In implementation, the normal range is stored in the system configuration file and supports dynamic updates. For example, the sleep duration range can be adjusted according to the user's age, and the range for the elderly may be set to 5-8 hours. The identification of triggering conditions is based on comparison algorithms, such as using conditional statements to compare values. For example, if the sleep onset time data is outside of 22:00-24:00, an abnormality is triggered. The principle of this step is to use preset benchmarks for automated judgment, avoiding subjective bias and improving the objectivity of the assessment.

[0062] Step S105: Record the data information corresponding to the recognition results according to the data type of sleep behavior data. The triggering conditions include the first triggering condition, the second triggering condition, the third triggering condition, the fourth triggering condition, and the fifth triggering condition.

[0063] Explained, the trigger conditions correspond one-to-one with the sleep behavior data types. The first trigger condition targets sleep onset time data, the second targets sleep duration data, the third targets the number of awakenings data, the fourth targets sleep efficiency data, and the fifth targets the proportion of deep sleep data. The system records the identification results based on the data type, including the anomaly type, actual value, normal range, and timestamp. For example, if the sleep onset time data is abnormal, the system will record "Abnormal sleep onset time, actual value 01:00, normal range 22:00-24:00". These records are stored in the database for subsequent analysis. In implementation, the system uses data structures, such as dictionaries or arrays, to map trigger conditions to specific data fields and stores the records through database tables. The principle of this step is structured data management, which ensures the traceability and consistency of abnormal information and provides detailed input for risk assessment.

[0064] Step S106: The first trigger condition is the time to fall asleep, the second trigger condition is the sleep duration, the third trigger condition is the number of times you wake up, the fourth trigger condition is the sleep efficiency, and the fifth trigger condition is the proportion of deep sleep.

[0065] Explained, these trigger conditions clearly define the target data for each anomaly detection, ensuring the clarity of the system logic. The first trigger condition focuses on sleep onset time data, checking if it falls within a preset range; the second trigger condition focuses on sleep duration data, checking if it is within a reasonable range; the third trigger condition focuses on wakefulness data, checking if it exceeds a threshold; the fourth trigger condition focuses on sleep efficiency data, checking if it is below the standard; and the fifth trigger condition focuses on deep sleep ratio data, checking if it deviates from normal. This division of labor is based on the independence of sleep parameters, with each condition handling a key aspect to avoid omissions. In implementation, the trigger conditions are implemented in the code as functions or rules. For example, the first trigger condition might be a function that takes sleep onset time data as input and outputs a boolean value (abnormal or normal). The principle behind this step is modular design, which improves the maintainability and scalability of the system.

[0066] The specific anomaly labeling methods include: responding to the first trigger condition result by comparing sleep onset time data with the sleep onset time interval; if the sleep onset time data is not within the sleep onset time interval, an anomaly is labeled; responding to the second trigger condition result by comparing sleep duration data with the sleep duration interval; if the sleep duration data is less than or greater than the sleep duration interval, an anomaly is labeled; responding to the third trigger condition result by comparing the number of awakenings data with the number of awakenings value; if the number of awakenings data is greater than the number of awakenings value; responding to the fourth trigger condition result by comparing sleep efficiency data with the sleep efficiency value; if the sleep efficiency data is less than the sleep efficiency value; and responding to the fifth trigger condition result by comparing the deep sleep ratio data with the deep sleep ratio value; if the deep sleep ratio data is less than or greater than the deep sleep ratio value, an anomaly is labeled.

[0067] Explained, the anomaly labeling process is based on comparison logic and threshold judgment principles. For the first trigger condition, the system compares the sleep onset time data with the sleep onset time range (22:00-24:00). If the data is outside this range, such as at 01:00, it is labeled as an anomaly, which may indicate a disrupted sleep schedule. For the second trigger condition, the system compares the sleep duration data with the sleep duration range (6-9 hours). If the data is less than 6 hours or greater than 9 hours, such as 5 hours or 10 hours, it is labeled as an anomaly, as both excessively short and long sleep durations can affect health. For the third trigger condition, the system compares the number of awakenings data with the wake-up count value (2 times). If the data is greater than 2 times, such as 3 times, it is labeled as an anomaly, as frequent awakenings disrupt sleep continuity. For the fourth trigger condition, the system compares... Sleep efficiency data is compared to a sleep efficiency value of 85%. If the data is less than 85%, such as 80%, it is marked as abnormal, indicating insufficient time utilization. For the fifth trigger condition, the system compares the deep sleep percentage data with the deep sleep percentage value of 20%-25%. If the data is less than 20% or greater than 25%, such as 15% or 30%, it is marked as abnormal, indicating insufficient or excessive deep sleep, which may reflect sleep quality problems. In implementation, the system uses conditional statements in programming languages ​​to perform the comparison and marks abnormal results as "abnormal," while recording detailed information (such as deviation values). The principle of this step is to achieve rapid anomaly detection through automated rules, reducing manual intervention and improving processing efficiency.

[0068] In this embodiment of the invention, this step is based on the principle of multi-source data acquisition and preset benchmark judgment. Through the collaborative work of a smart bracelet and a sleep diary application, comprehensive acquisition of sleep behavior data and automatic identification of anomalies are achieved. The smart bracelet uses sensor technology to capture objective physiological parameters, while the sleep diary application obtains subjective sleep experiences through human-computer interaction. This multi-source acquisition method is based on the principle of data fusion, effectively reducing the bias of a single data source. The preset normal range is based on medical research settings, ensuring the scientific nature of the assessment standards. The triggering conditions and anomaly labeling mechanism are based on comparison logic and threshold judgment, enabling rapid identification of data anomalies. The continuity of the steps is reflected in the coherent execution of raw data acquisition, normal range comparison, and anomaly labeling, providing an accurate and consistent data foundation for subsequent risk assessment. Through multi-source data acquisition and automated anomaly labeling, this step overcomes the problems of strong subjectivity and incomplete data in traditional insomnia assessment, providing high-quality, multi-dimensional input data. This enhances the reliability of insomnia risk assessment because multi-instrument collaboration can comprehensively capture sleep characteristics, while automated processing ensures the objectivity and efficiency of the assessment, thus laying a data foundation for the overall method and reducing the risk of misjudgment in subsequent analysis.

[0069] In one embodiment, S2, based on the number of marked anomalies, the system enters the evaluation node to obtain the insomnia risk level.

[0070] Explained, this step corresponds to step S2 of claim 1 and the content of claim 5. It involves risk assessment based on the number of anomaly labels. The purpose is to automatically determine the user's insomnia risk level by quantifying anomalies. The assessment node is a processing module in the system responsible for counting the number of anomaly labels and mapping them to risk levels according to preset thresholds. Specifically, the system queries the current sleep cycle from step S1, for example, the number of anomaly labels per night. If the user has abnormal sleep onset time and abnormal sleep efficiency, the number of anomalies is 2. The preset insomnia levels include low risk, medium risk, and high risk, and the risk level is based on the number of anomalies. An example of threshold setting is as follows: low risk is defined as the number of anomalies being less than or equal to a first threshold. For example, the first threshold is set to... 1 indicates 0 or 1 anomaly; medium risk means the number of anomalies is greater than the first threshold and less than or equal to the second threshold. For example, if the second threshold is set to 3, it means 2 or 3 anomalies; high risk means the number of anomalies is greater than the second threshold. For example, more than 3 anomalies. The assessment process uses a counter to count the number of anomalies and then compares it with the threshold. For example, if the number of anomalies is 2, it is assessed as medium risk. The assessment results are stored in the user profile. The threshold can be adjusted according to the actual application scenario. For example, in strict mode, the first threshold may be set to 0 and the second threshold may be set to 2. The logic of the assessment node is implemented through simple arithmetic comparison, which ensures efficiency. The principle of this step is quantitative analysis, which aggregates discrete abnormal events into a comprehensive risk indicator, avoiding overreaction to a single anomaly.

[0071] In this embodiment of the invention, this step is based on the principles of quantitative analysis and threshold judgment. By counting the number of anomaly labels and mapping them to preset risk levels, an automated assessment of insomnia risk is achieved. The assessment node uses a counter to count the number of anomalies and classifies them by comparing them with a threshold. This quantitative method, based on statistical principles, ensures the objectivity and consistency of the assessment. The threshold setting can be adjusted according to actual needs, enhancing the system's adaptability. After statistical comparison, the risk level is output, providing a clear basis for subsequent graded interventions. By quantifying the number of anomalies for risk assessment, this step solves the problems of relying on subjective experience and lacking unified standards in traditional methods, achieving rapid and objective risk level determination. This improves the efficiency of insomnia management because automated assessment reduces labor costs, and clear level classification facilitates targeted interventions, thereby optimizing resource allocation and improving the overall treatment effect.

[0072] In one embodiment, S3, based on the insomnia risk level, a corresponding processing procedure is triggered, and an analysis report is output after the processing procedure is completed.

[0073] Explainedly, this step corresponds to step S3 of claim 1 and the content of claims 6-9. It involves implementing graded interventions based on risk levels, with the aim of addressing different levels of insomnia risk through different processing procedures. The final step is to generate an analysis report to summarize the assessment and intervention results. This step includes three sub-processes: routine monitoring, early warning intervention, and expert consultation. Each sub-process is designed for a specific risk level to ensure the targetedness and effectiveness of the intervention.

[0074] Step S301: When the insomnia risk level is low, proceed with the routine monitoring process, including adjusting the normal sleep range and updating the insomnia risk level.

[0075] Explained, the routine monitoring process is applicable to low-risk users. Its main purpose is to maintain users' sleep health through continuous monitoring and personalized adjustments. This process is preventative and focuses on long-term tracking. Specifically, it includes collecting basic user information, such as name, age, gender, and medical history. This information is obtained through user registration or questionnaires. For example, age affects sleep needs; older adults may have shorter sleep durations. Medical history, such as anxiety disorders, may reveal underlying causes of sleep problems. Based on the basic information and sleep behavior data, a personalized monitoring plan is developed. This plan includes identifying abnormal sleep behavior data, obtaining historical sleep behavior data corresponding to the identified abnormal sleep behavior data, and adjusting the normal sleep range based on the historical sleep behavior data. The historical sleep behavior data refers to sleep behavior over the 30 days prior to the current acquisition time stamp. The data includes details of the personalized monitoring plan, identifying and labeling abnormal sleep behavior data. The system identifies which data types are labeled as abnormal; for example, if a user frequently has abnormal sleep onset time, the system will focus on sleep onset time data. Historical sleep behavior data is obtained; the system extracts relevant data from the database for the past 30 days, such as the average and standard deviation of daily sleep onset time. The system adjusts the normal sleep range based on historical data to calculate the user's personalized normal range. For example, if a user's historical average sleep onset time is 23:30 with a standard deviation of 30 minutes, the system may adjust the sleep onset time range to 23:00 to 24:00. Similarly, the sleep duration range may be adjusted based on historical averages, such as 7-8 hours. The adjustment algorithm uses statistical methods, such as moving averages or percentile calculations; for example, the 25th and 75th percentiles of sleep duration over the past 30 days are calculated as the new range. After adjustment, the system reassesses the insomnia risk level and updates the user profile. Routine monitoring procedures are executed regularly, for example, weekly, to ensure adaptability. The principle behind this step is adaptive learning, which dynamically optimizes the benchmark using historical data to improve the personalization of the evaluation.

[0076] Step S302: When the insomnia risk level is medium risk, the early warning intervention process is initiated, including sending early warning information to the user and providing improvement suggestions.

[0077] Interpretationally, the early warning intervention process is applicable to medium-risk users and aims to prevent the risk of insomnia from escalating further through predictive and early warning mechanisms. This process is proactive and focuses on risk mitigation. Specifically, it involves establishing a correlation between pre-stored high-risk sleep behavior data and time periods; predicting future insomnia risk using historical sleep behavior data and this correlation; obtaining prediction results; if the prediction result exceeds a sixth threshold, increasing the insomnia risk level and issuing a warning; providing users with targeted improvement suggestions through a human-computer interaction interface; the system stores patterns of high-risk sleep behavior data, such as increased awakenings over three consecutive days or a declining trend in sleep efficiency, based on clinical research or historical data analysis; a prediction model using time series analysis or machine learning algorithms, such as the ARIMA model or recurrent neural networks, analyzes historical and current sleep behavior data. Model inputs include historical data, such as sleep efficiency over the past 30 days, and current data, such as the number of abnormalities in the most recent night; the output is a future risk value, such as the probability of insomnia risk within the next 7 days; the prediction result is a numerical value representing the risk level or probability, for example, a prediction range of 0-1, with 0.7 indicating a high-risk probability. The sixth threshold is a preset value; for example, 0.6 represents a 60% probability. If the predicted result exceeds the threshold, the system temporarily upgrades the user's risk level, for example, from medium risk to high risk, and sends an early warning notification via push notification, SMS, or email. The warning information includes a risk description and emergency action suggestions. Targeted improvement suggestions are provided to users through a human-computer interaction interface. These suggestions are based on the specific type of anomaly. For example, if the sleep onset time is abnormal, it is recommended to establish a fixed sleep schedule and avoid evening caffeine intake; if the number of awakenings is abnormal, it is recommended to optimize the bedroom environment, such as reducing noise and engaging in relaxation exercises; if sleep efficiency is abnormal, it is recommended to limit time spent in bed and increase daytime physical activity. The human-computer interaction interface can be a notification center in a mobile application or a web portal, ensuring easy access for users. The principle behind this step lies in predictive analytics and personalized feedback, using models to predict risk trends and providing customized suggestions based on the user's specific problems.

[0078] Step S303: When the insomnia risk level is high, the expert consultation process is initiated, which includes automatically matching relevant experts and pushing assessment reports.

[0079] Explained, the expert consultation process is designed for high-risk users, providing professional-level intervention and support. This responsive process focuses on addressing severe sleep problems. Specifically, it involves acquiring corresponding sleep behavior data based on anomaly labeling, determining the insomnia risk type based on this data, matching the insomnia risk type with appropriate experts, and generating an analysis report of key sleep indicators. Insomnia risk types include difficulty falling asleep, difficulty maintaining sleep, early awakening, low sleep efficiency, and insufficient deep sleep. The analysis report includes basic information, historical sleep behavior data, improvement suggestions, and doctor information. The determination of insomnia risk type is based on the pattern of anomaly labeling; for example, difficulty falling asleep is primarily characterized by abnormal sleep onset time, and difficulty maintaining sleep is primarily characterized by early awakening. The system identifies several sleep risk types, including: abnormal frequency of awakenings; early awakenings characterized by excessively short sleep duration (e.g., waking up earlier than expected); low sleep efficiency characterized by abnormal sleep efficiency; and insufficient deep sleep characterized by an abnormal proportion of deep sleep. The classification logic prioritizes the type with the most abnormalities, or uses a rule engine (e.g., decision tree) to classify based on multiple abnormal combinations. Expert matching involves maintaining an expert database, including experts' areas of expertise (e.g., specializing in treating difficulty falling asleep or sleep maintenance), contact information, and working hours. Based on the insomnia risk type, the system automatically queries the database to match the most suitable expert; for example, for users with difficulty falling asleep, an expert specializing in cognitive behavioral therapy is matched. An analysis report is generated, including basic information such as user name, age, and gender; historical sleep behavior data (trend charts of sleep onset time and sleep efficiency over the past 30 days); improvement suggestions (preliminary suggestions generated by the system, such as those extracted from early warning intervention processes); and doctor information (name, professional background, and contact information of the matched expert). The report is pushed to the matched expert via a secure channel, such as encrypted email or a dedicated platform, and the user is notified. Experts can then conduct a preliminary diagnosis or arrange consultation based on the report. The principle behind this step is the integration of professional resources and automated matching, which improves intervention efficiency through systematic processing.

[0080] In this embodiment of the invention, this step is based on the principles of tiered intervention and adaptive processing. By triggering corresponding processing procedures for different risk levels, it achieves precise management and effective intervention of insomnia risk. The routine monitoring process utilizes the principle of adaptive learning to dynamically adjust the normal range based on historical data, improving the personalization of monitoring. The early warning intervention process is based on predictive analysis, using time series models to predict risk trends and achieve early warning. The expert consultation process integrates professional resources, providing professional-level support through automated matching and report generation. After process judgment, the corresponding sub-processes are executed, and finally, an analysis report is output to form a closed loop. Through the tiered processing process, this step overcomes the problems of one-size-fits-all and delayed response in traditional interventions, achieving full-chain adaptive management from routine monitoring to professional consultation. This significantly improves the effectiveness of insomnia intervention and user experience because tiered processing ensures the rational allocation of resources, the adaptive mechanism improves the accuracy of intervention, and automated execution reduces the human burden, thereby optimizing the overall efficiency of sleep health management.

[0081] Example 2, refer to Figure 2 It provides a data processing system based on insomnia behavior, including a data collection module, an assessment module, and a processing module. These modules can be implemented through a combination of software and hardware. For example, the system is deployed on a cloud server and communicates with user devices (smart bracelets, smartphones) via a network.

[0082] The data acquisition module is used to collect users' sleep behavior data based on different instruments. Based on the sleep behavior data, it receives trigger conditions and marks anomalies. The data acquisition module includes a data interface submodule, which processes the data input of smart bracelets and sleep diary applications; a storage submodule, which stores raw data and anomaly records in a database; and a processing submodule, which executes the anomaly marking logic.

[0083] The assessment module is used to enter the assessment node based on the number of anomalies and obtain the insomnia risk level. The assessment module includes a counting submodule to count the number of anomalies, a threshold comparison submodule to compare the number of anomalies with the threshold, and a mapping submodule to output the risk level.

[0084] The processing module is used to trigger corresponding processing procedures based on the risk level of insomnia. Once the processing procedure is completed, an analysis report is output. The processing module includes a process control sub-module, which activates routine monitoring, early warning intervention, or expert consultation sub-modules based on the risk level, and a report generation sub-module, which creates the analysis report.

[0085] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0086] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A data processing method based on insomnia behavior, characterized in that, Includes the following steps: Step S1: Collect sleep behavior data of users using different instruments, and receive trigger conditions and mark abnormalities based on the sleep behavior data; Step S2: Based on the number of marked abnormalities, proceed to the assessment node to obtain the insomnia risk level; Step S3: Based on the insomnia risk level, trigger the corresponding processing flow, and output an analysis report after the processing flow is completed.

2. The data processing method based on insomnia behavior as described in claim 1, characterized in that, Step S1 includes the following sub-steps: Step S101: Collect sleep behavior data of users according to different instruments. The sleep behavior data includes sleep onset time data, sleep duration data, number of awakenings data, sleep efficiency data, and deep sleep ratio data. Step S102: Obtain the sleep onset time data, sleep duration data, number of awakenings data, and deep sleep ratio data through the smart bracelet; Step S103: Obtain sleep efficiency data through a sleep diary application.

3. The data processing method based on insomnia behavior as described in claim 2, characterized in that, Step S1 further includes the following sub-steps: Step S104: Based on the sleep behavior data, a normal sleep range is preset; based on the normal sleep range, trigger conditions are received; the trigger conditions are identified; and identification results are obtained. The normal sleep range includes the sleep onset time range, sleep duration range, number of awakenings, sleep efficiency value, and deep sleep ratio value. Step S105, and record the data information corresponding to the recognition result according to the data type of sleep behavior data. The triggering conditions include a first triggering condition, a second triggering condition, a third triggering condition, a fourth triggering condition, and a fifth triggering condition. Step S106: The first trigger condition is sleep onset time data, the second trigger condition is sleep duration data, the third trigger condition is wake-up frequency data, the fourth trigger condition is sleep efficiency data, and the fifth trigger condition is deep sleep ratio data.

4. The data processing method based on insomnia behavior as described in claim 3, characterized in that, The anomalies specifically include: In response to the first trigger condition result, the sleep time data is compared with the sleep time interval. If the sleep time data is not within the sleep time interval, it is marked as abnormal. In response to the second trigger condition result, the sleep duration data is compared with the sleep duration range. When the sleep duration data is less than or greater than the sleep duration range, an anomaly is marked. In response to the third trigger condition result, the number of times the user wakes up is compared with the number of times the user wakes up. If the number of times the user wakes up is greater than the number of times the user wakes up, an anomaly is marked. In response to the fourth trigger condition result, the sleep efficiency data is compared with the sleep efficiency value. If the sleep efficiency data is less than the sleep efficiency value, an anomaly is marked. In response to the fifth trigger condition result, the deep sleep ratio data is compared with the deep sleep ratio value. When the deep sleep ratio data is less than or greater than the deep sleep ratio value, an anomaly is marked.

5. The data processing method based on insomnia behavior as described in claim 4, characterized in that, Step S2 specifically includes: Based on the anomalies, an assessment node is established to obtain the insomnia risk level. The assessment node involves obtaining the number of anomalies based on the anomalies, and then assessing the number of anomalies against a preset insomnia level to obtain the insomnia risk level. The preset insomnia level includes low risk, medium risk, and high risk. The low risk is defined as the number being less than or equal to a first threshold. The medium risk is defined as the number of items being greater than a first threshold and less than or equal to a second threshold; The high risk is defined as the number of [number] items being greater than the second threshold.

6. The data processing method based on insomnia behavior as described in claim 5, characterized in that, Step S3 includes the following sub-steps: Step S301: When the insomnia risk level is low, proceed with the routine monitoring process, including adjusting the normal sleep range and updating the insomnia risk level. Step S302: When the insomnia risk level is medium risk, the early warning intervention process is initiated, including sending early warning information to the user and providing improvement suggestions. Step S303: When the insomnia risk level is high, the expert consultation process is initiated, including automatically matching relevant experts and pushing assessment reports.

7. The data processing method based on insomnia behavior as described in claim 6, characterized in that, The routine monitoring process specifically includes: Collect basic information about the user, including name, age, gender, and medical history; Based on the basic information and sleep behavior data, a personalized monitoring plan is formulated. The personalized monitoring plan includes identifying abnormal sleep behavior data based on the sleep behavior data, obtaining historical sleep behavior data corresponding to the abnormal sleep behavior data, and adjusting the normal sleep range based on the historical sleep behavior data. The historical sleep behavior data is the sleep behavior data of the 30 days prior to the current timestamp of the acquired sleep behavior data.

8. The data processing method based on insomnia behavior as described in claim 7, characterized in that, The early warning and intervention process specifically includes: A correlation is established based on pre-stored high-risk sleep behavior data and time. By using the historical sleep behavior data, sleep behavior data and the correlation, the risk of future insomnia is predicted, and the prediction results are obtained. If the predicted result is greater than the sixth threshold, the insomnia risk level is increased and an early warning message is issued; Provide users with targeted improvement suggestions through a human-computer interaction interface.

9. The data processing method based on insomnia behavior as described in claim 8, characterized in that, The expert consultation process specifically includes: Based on the marked anomalies, corresponding sleep behavior data is obtained. Based on the sleep behavior data, the insomnia risk type is obtained. Based on the insomnia risk type, a corresponding expert is matched, and an analysis report of key sleep indicators is generated. The insomnia risk types include difficulty falling asleep, difficulty maintaining sleep, early awakening, low sleep efficiency, and insufficient deep sleep. The analysis report includes basic information, historical sleep behavior data, improvement suggestions, and doctor information.

10. A data processing system based on insomnia behavior, applied in the data processing method based on insomnia behavior as described in any one of claims 1-9, characterized in that, It includes a data acquisition module, an evaluation module, and a processing module; The acquisition module is used to collect users' sleep behavior data according to different instruments, and to receive trigger conditions and mark abnormalities based on the sleep behavior data; The assessment module is used to enter the assessment node based on the number of marked abnormalities and obtain the insomnia risk level; The processing module is used to trigger a corresponding processing flow based on the insomnia risk level, and output an analysis report after the processing flow is completed.

Images