Intelligent bed sleep quality analysis method and system

Abstract

The invention relates to the field of smart home, and provides a smart bed sleep quality analysis method, which comprises the following steps: firstly, determining a sleep monitoring area of a smart bed, obtaining environment information, querying sleeper characteristics according to the environment information, determining target sleep data, performing segmentation processing, calculating a sleep fluctuation value and a characteristic deviation degree, and correcting to obtain a sleep correction parameter; querying key correction indexes, performing trend marking, dividing continuous association time periods, updating to obtain trend updating indexes, mining index association characteristics, constructing a sleep quality analysis strategy in combination with life data of a sleeper, performing simulation analysis to obtain strategy simulation scores, and finally generating a sleep quality analysis report corresponding to the intelligent bed according to the strategy simulation scores. The analysis effect of the sleep quality of the intelligent bed can be improved.

Description

Technical Field

[0001] The present invention relates to the field of smart homes, and in particular to a smart bed sleep quality analysis method and system. Background Art

[0002] In the field of smart home, good sleep quality is crucial to people's physical and mental health. It is closely related to people's daily work efficiency, emotional state, physical immunity and other aspects. In terms of sleep quality monitoring and analysis, the smart bed sleep quality analysis method is key to accurately assessing sleep conditions and providing personalized sleep recommendations.

[0003] At present, traditional sleep quality assessment mainly relies on subjective feelings and some simple wearable devices, such as bracelets. This method has exposed many shortcomings in the face of the growing demand for accurate sleep monitoring and complex and diverse individual sleep differences. On the one hand, subjective feelings are highly subjective and uncertain, and it is difficult to fully and objectively reflect the real sleep quality. On the other hand, although traditional wearable devices can collect some sleep data, the data dimensions are often limited. For example, bracelets mainly focus on monitoring heart rate, movement and other data, and lack multi-dimensional information such as sleeping posture changes, mattress pressure distribution, and breathing patterns during sleep. It is impossible to conduct in-depth and accurate analysis of sleep quality and it is difficult to meet the needs of personalized sleep improvement. Therefore, a smart bed sleep quality analysis method is needed to improve the analysis effect of smart bed sleep quality. Summary of the invention

[0004] The present invention provides a smart bed sleep quality analysis method and system, the main purpose of which is to improve the analysis effect of the smart bed sleep quality.

[0005] To achieve the above object, the present invention provides a method for analyzing sleep quality of a smart bed, comprising:

[0006] Determine a sleep monitoring area corresponding to the smart bed, obtain sleep environment information corresponding to the sleep monitoring area, and query sleeper characteristics corresponding to a sleeper in the sleep monitoring area according to the sleep environment information;

[0007] Based on the sleep environment information and the sleeper characteristics, determining target sleep data corresponding to the smart bed, segmenting the target sleep data to obtain multiple sleep data segments, and calculating sleep fluctuation values ​​corresponding to the multiple sleep data segments;

[0008] Identify the characteristic index numbers corresponding to the multiple sleep data, calculate the characteristic deviations corresponding to the multiple sleep data based on the characteristic index numbers, and perform correction processing on the multiple sleep data based on the sleep fluctuation values ​​and the characteristic deviations to obtain sleep correction parameters;

[0009] Querying the key correction index in the sleep correction parameter, and trend-marking the key correction index to obtain a trend-marked index, dividing the continuous association time period corresponding to the trend-marked index, and trend-updating the trend-marked index based on the continuous association time period to obtain a trend update index;

[0010] The indicator correlation characteristics between the trend update indicators are mined, and a sleep quality analysis strategy corresponding to the smart bed is constructed based on the indicator correlation characteristics and the preset life data of the sleeper. The sleep quality analysis strategy is simulated and analyzed to obtain a strategy simulation score, and a sleep quality analysis report corresponding to the smart bed is generated based on the strategy simulation score.

[0011] Optionally, querying the sleeper characteristics corresponding to the sleeper in the sleep monitoring area according to the sleep environment information includes:

[0012] Determining the information data range in the sleeping environment information;

[0013] Divide the environmental comfort level corresponding to the information data range;

[0014] Calculating the environmental tolerance of the sleeper in the sleep monitoring area based on the environmental comfort level;

[0015] Based on the environmental tolerance, analyzing the environmental reaction tendency of the sleeper in the sleep monitoring area;

[0016] Based on the environmental reaction tendency, the sleeper characteristics corresponding to the sleeper in the sleep monitoring area are queried.

[0017] Optionally, the segmenting of the target sleep data to obtain multiple segments of sleep data includes:

[0018] Analyzing the data time axis corresponding to the target sleep data;

[0019] Based on the data time axis, determining a sleep cycle interval corresponding to the target sleep data;

[0020] Analyzing the rapid eye movement period in the sleep cycle interval;

[0021] Identifying period boundary features corresponding to the rapid eye movement period;

[0022] Separating the non-rapid eye movement period in the sleep cycle interval based on the cycle boundary feature;

[0023] Based on the rapid eye movement period and the non-rapid eye movement period, the target sleep data is segmented to obtain multiple sleep data.

[0024] Optionally, calculating the sleep fluctuation values ​​corresponding to the multiple sleep data segments includes:

[0025] The sleep fluctuation value corresponding to the multiple sleep data is calculated using the following formula:

[0026]

[0027] Wherein, V represents the sleep fluctuation value corresponding to the multiple sleep data segments, M represents the number of segments into which the sleep data is divided, and t j represents the duration of the jth sleep period, δ j represents the standard deviation of sleep physiological indicators in the jth sleep segment, μ j represents the average value of sleep physiological indicators in the jth sleep segment, It represents the overall average value of sleep physiological indicators in all sleep segments.

[0028] Optionally, the calculating, based on the characteristic index number, characteristic deviations corresponding to the plurality of sleep data segments includes:

[0029] The characteristic deviation corresponding to the multiple sleep data segments is calculated using the following formula:

[0030]

[0031] Wherein, D represents the characteristic deviation corresponding to the plurality of sleep data segments, N represents the total number of sleep data segments, i represents the number index of sleep data segments, Q represents the total number of characteristic indicators, p represents the number index of characteristic indicators, and x represents the total number of characteristic indicators. ip represents the actual value of the index corresponding to the pth characteristic index in the i-th sleep data. Represents the average value of the index corresponding to the pth feature index in all sleep data segments.

[0032] Optionally, the correcting the plurality of sleep data segments based on the sleep fluctuation value and the characteristic deviation to obtain a sleep correction parameter includes:

[0033] Determining a fluctuation deviation coefficient corresponding to each sleep data segment based on the sleep fluctuation value and the characteristic deviation;

[0034] Normalizing the fluctuation deviation coefficient to obtain a normalized deviation value;

[0035] Querying a correction weight corresponding to the normalized deviation value;

[0036] Performing weighted processing on the plurality of sleep data segments according to the correction weights to obtain weighted data values;

[0037] Based on the weighted data value, parameter extraction is performed on the weighted sum value to obtain a sleep correction parameter.

[0038] Optionally, the step of performing trend marking on the key correction indicator to obtain a trend marking indicator includes:

[0039] Constructing an indicator marking rule corresponding to the key correction indicator;

[0040] Based on the indicator marking rule, collecting the indicator data sequence corresponding to the key correction indicator;

[0041] Extracting sequence change characteristics corresponding to the indicator data sequence;

[0042] Analyze the change trend corresponding to the data change characteristics;

[0043] Based on the change trend, the key correction indicator is trend-marked to obtain a trend-marked indicator.

[0044] Optionally, the updating of the trend marking indicator based on the continuous association period to obtain the trend update indicator includes:

[0045] Determining initial change characteristics corresponding to the trend marking indicator within the continuous correlation period;

[0046] Analyzing the change characteristics corresponding to the initial change features;

[0047] Based on the change characteristics, key trend segments in the trend marking indicator are screened out;

[0048] Based on the key trend segmentation, construct a trend update framework corresponding to the trend marking indicator;

[0049] Based on the trend update framework, the trend marking indicator is trend updated to obtain a trend update indicator.

[0050] Optionally, the sleep quality analysis strategy corresponding to the smart bed is constructed based on the indicator association characteristics in combination with preset life data of the sleeper, including:

[0051] Query the correlation impact factor corresponding to the indicator correlation characteristic;

[0052] Collecting life detail data related to the associated influencing factors and preset life data of the sleeper;

[0053] Analyzing the correlation mapping between the life detail data and the indicator correlation characteristics;

[0054] Based on the association mapping, determining the core quality options corresponding to the smart bed;

[0055] Based on the core quality options, a sleep quality analysis strategy corresponding to the smart bed is constructed.

[0056] Optionally, in order to solve the above problem, the present invention provides a smart bed sleep quality analysis system, the system comprising:

[0057] A feature query module, used to determine the sleep monitoring area corresponding to the smart bed, obtain the sleep environment information corresponding to the sleep monitoring area, and query the sleeper characteristics corresponding to the sleeper in the sleep monitoring area according to the sleep environment information;

[0058] A fluctuation value calculation module, used to determine the target sleep data corresponding to the smart bed based on the sleep environment information and the sleeper characteristics, segment the target sleep data to obtain multiple sleep data, and calculate the sleep fluctuation values ​​corresponding to the multiple sleep data;

[0059] a calibration processing module, configured to identify characteristic index numbers corresponding to the plurality of sleep data segments, calculate characteristic deviations corresponding to the plurality of sleep data segments based on the characteristic index numbers, and perform correction processing on the plurality of sleep data segments based on the sleep fluctuation values ​​and the characteristic deviations to obtain sleep correction parameters;

[0060] a trend update module, used for querying the key correction index in the sleep correction parameter, and trend-marking the key correction index to obtain a trend-marked index, dividing the continuous association time period corresponding to the trend-marked index, and trend-updating the trend-marked index based on the continuous association time period to obtain a trend update index;

[0061] A report generation module is used to mine the indicator correlation characteristics between the trend update indicators, construct a sleep quality analysis strategy corresponding to the smart bed based on the indicator correlation characteristics and preset sleeper life data, simulate and analyze the sleep quality analysis strategy to obtain a strategy simulation score, and generate a sleep quality analysis report corresponding to the smart bed based on the strategy simulation score.

[0062] First of all, by determining the sleep monitoring area corresponding to the smart bed, the present invention can focus on the core range of data collection, avoid data interference from irrelevant areas, and can be more reasonably distributed and configured. While ensuring comprehensive monitoring, it reduces costs and energy consumption, improves the overall performance of the equipment, and thus greatly improves the accuracy and effectiveness of the acquired sleep data. At the same time, based on the sleep environment information and the characteristics of the sleeper, the present invention determines the target sleep data corresponding to the smart bed, and can accurately locate key data. For example, based on the sleeper's sensitivity to noise, it focuses on data for related time periods to improve data pertinence, which helps to deeply analyze the factors affecting sleep quality, and provides an accurate basis for personalized sleep improvement strategies in combination with ambient temperature and the sleeper's physique, thereby effectively improving the accuracy and effectiveness of sleep quality analysis. By identifying the characteristic index numbers corresponding to the multiple sleep data segments, the present invention can quantify sleep quality, which helps to accurately evaluate sleep quality. The characteristic index numbers can provide a more comprehensive understanding of the conditions of each stage of sleep, such as light sleep, deep sleep, and so on. The characteristic performance of sleep can provide a quantitative basis for the diagnosis of sleep disorders. When abnormal sleep is found, the specific characteristic index number can assist in determining the problem, and then provide a reliable basis for formulating strategies to improve sleep. The present invention can accurately locate the core problem area in the sleep data by querying the key correction index in the sleep correction parameter, and can quickly focus on the key factors affecting sleep quality. By clarifying these key indicators, it can efficiently provide a direct basis for personalized sleep improvement plans, which helps to adjust sleep habits, environment, etc. in a targeted manner, thereby effectively improving sleep quality and preventing potential sleep disorders. Further, the present invention helps to deeply understand the intrinsic connection between various elements of sleep by mining the indicator correlation characteristics between the trend update indicators, extract systematic laws from complex sleep data, and more accurately grasp the collaborative operation mode of the sleep mechanism, providing a key perspective for a comprehensive and in-depth understanding of the sleep state, and then providing a strong basis for the formulation and optimization of personalized sleep improvement plans. Therefore, the sleep quality analysis method and system of a smart bed proposed by the present invention can improve the analysis effect of the sleep quality of the smart bed. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 A schematic diagram of a process for analyzing sleep quality of a smart bed provided by an embodiment of the present invention;

[0064] Figure 2 A schematic diagram of modules for implementing a smart bed sleep quality analysis system provided in one embodiment of the present invention.

[0065] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0066] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0067] The embodiment of the present application provides a method for analyzing the sleep quality of a smart bed. The execution subject of the method for analyzing the sleep quality of a smart bed includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for analyzing the sleep quality of a smart bed can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0068] Embodiment 1:

[0069] Reference Figure 1 , which is a flow chart of a method for analyzing sleep quality of a smart bed provided in one embodiment of the present invention.

[0070] In this embodiment, the smart bed sleep quality analysis method includes:

[0071] S1. Determine a sleep monitoring area corresponding to the smart bed, obtain sleep environment information corresponding to the sleep monitoring area, and query sleeper characteristics corresponding to a sleeper in the sleep monitoring area based on the sleep environment information.

[0072] By determining the sleep monitoring area corresponding to the smart bed, the present invention can focus on the core range of data collection, avoid data interference from irrelevant areas, and can be distributed and configured more reasonably. While ensuring comprehensive monitoring, it reduces costs and energy consumption, improves the overall performance of the equipment, and thus greatly improves the accuracy and effectiveness of the acquired sleep data.

[0073] Among them, the smart bed refers to a bed device equipped with advanced sensing technology, a data processing unit and intelligent interactive functions. It is not just an ordinary bed for people to rest and sleep, but can also perceive various physiological parameters and behavioral movements of the human body during sleep in real time, and collect, analyze and process this information through a built-in intelligent system. For example, the smart bed may have built-in pressure sensors, acceleration sensors, heart rate monitors and other types of sensors. These sensors work together to capture the subtle changes of the human body during sleep in all directions; the sleep monitoring area refers to a specific range or area on the smart bed specifically used to monitor human sleep-related data, usually covering the main part of the human body that contacts the mattress when sleeping, such as the mattress area corresponding to key parts such as the head, neck, back, buttocks, legs, etc. Optionally, the determination of the sleep monitoring area corresponding to the smart bed can be achieved through a sensor array layout method, such as: using the sensor array built into the smart bed to determine the sleep monitoring area.

[0074] Furthermore, by acquiring the sleep environment information corresponding to the sleep monitoring area, the present invention can accurately grasp the impact of environmental factors such as temperature, humidity, light intensity, noise level, etc. on sleep, provide a basis for improving the sleep environment, help create comfortable sleeping conditions, and help analyze the relationship between the environment and the sleeper's sleep quality fluctuations, thereby providing a key reference for the formulation of personalized sleep strategies and improving the pertinence and effectiveness of overall sleep quality improvement.

[0075] Among them, the sleeping environment information refers to the sum of a series of external condition data that have a direct or indirect impact on the sleep quality of the sleeper within the sleep monitoring area of ​​the smart bed, including but not limited to the temperature and humidity conditions of the environment. The suitability of the two will affect the sleeper's physical comfort and sleep depth; the brightness and color of the light. Too strong or inappropriate light will interfere with the sleep rhythm; the volume and frequency of the noise. A noisy environment can easily lead to sleep interruptions and increased shallow sleep; air quality such as oxygen content, concentration of harmful gases, etc. Optionally, the acquisition of the sleeping environment information corresponding to the sleep monitoring area can be achieved through sensor monitoring equipment, such as: HIH-4000, TSL2561.

[0076] Furthermore, the present invention queries the sleeper characteristics corresponding to the sleepers in the sleep monitoring area based on the sleep environment information, and can accurately associate environmental factors with the sleeper characteristics. For example, it can analyze the differences in sleep responses of sleepers with different physiques in a high temperature environment, thereby providing a basis for personalized sleep improvement. On the other hand, it helps to deeply understand the impact of the environment on specific groups of sleepers with characteristics, such as the study of sleep changes in elderly sleepers in a noisy environment, so as to optimize the sleep environment settings in a targeted manner and improve the overall sleep quality and health level.

[0077] Among them, the sleeper characteristics refer to personal traits related to sleep, including but not limited to sleeping habits (such as whether there is a habit of taking naps, sleeping posture preferences, etc.), sleep quality fluctuation patterns (such as changes in sleep quality in specific seasons or environments), potential sleep disorders (such as whether you are prone to snoring, whether you have insomnia tendencies, etc.) and personalized needs for the sleeping environment (such as preference for harder or softer mattresses, preference for aromatherapy to aid sleep, etc.).

[0078] As an embodiment of the present invention, querying the sleeper characteristics corresponding to the sleeper in the sleep monitoring area based on the sleep environment information includes: determining the information data range in the sleep environment information; dividing the environmental comfort level corresponding to the information data range; calculating the environmental tolerance corresponding to the sleeper in the sleep monitoring area based on the environmental comfort level; analyzing the environmental reaction tendency corresponding to the sleeper in the sleep monitoring area based on the environmental tolerance; and querying the sleeper characteristics corresponding to the sleeper in the sleep monitoring area based on the environmental reaction tendency.

[0079] Among them, the information data range refers to the value range of various data in the sleep environment information, such as the high and low temperature range, the humidity range, the light brightness value and the decibel value of the noise; the environmental comfort level refers to the level indicating the environmental comfort level determined after classifying the information data range according to specific standards. For example, the temperature between 22-26 degrees Celsius, the humidity between 40%-60%, the light is dim and the noise is less than 30 decibels. It can be classified as a high comfort level, and the comfort level will decrease if it deviates from this range; the environmental tolerance refers to a quantitative assessment of the sleeper's ability to withstand and adapt to different environmental comfort levels. For example, a sleeper who can still maintain a good sleep state in a high temperature and high humidity environment has a higher environmental tolerance, and vice versa; the environmental reaction tendency refers to the trends and characteristics of the sleeper's physical and psychological reactions when facing changes in different environmental comfort levels, such as increased tossing frequency, sleep interruption, and emotional irritability.

[0080] Furthermore, the determination of the range of the information data in the sleep environment information can be achieved through a range determination method, such as: setting up a data acquisition terminal in the sleep monitoring area, connecting various environmental sensors, collecting data at certain time intervals, and then transmitting the data to a database for storage and analysis, so as to determine the range of the information data; the division of the environmental comfort level corresponding to the information data range can be achieved through a fuzzy logic algorithm, such as: based on existing experience and research, defining fuzzy sets of environmental factors such as temperature, humidity, light and noise, such as defining fuzzy sets such as "cold", "suitable" and "hot" for temperature, and then dividing the environmental comfort level according to the fuzzy set to which the information data range belongs; the calculation Calculating the environmental tolerance corresponding to the sleeper in the sleep monitoring area can be achieved through a machine learning algorithm, such as a decision tree algorithm, etc.; analyzing the environmental reaction tendency corresponding to the sleeper in the sleep monitoring area can be achieved through a behavioral observation method, such as using a camera installed in the sleep area (such as a camera with an infrared function) and using computer vision technology (such as the OpenCV library) to observe the behavior of the sleeper, and analyzing the sleeper's turning over frequency, limb movements and other behavioral changes when environmental factors change, so as to determine the environmental reaction tendency; querying the sleeper's corresponding sleeper characteristics in the sleep monitoring area can be achieved through a feature query tool, such as NexRing, C-Life and other tools.

[0081] S2. Based on the sleep environment information and the sleeper characteristics, determine the target sleep data corresponding to the smart bed, segment the target sleep data to obtain multiple sleep data segments, and calculate the sleep fluctuation values ​​corresponding to the multiple sleep data segments.

[0082] The present invention determines the target sleep data corresponding to the smart bed based on the sleep environment information and the characteristics of the sleeper, and can accurately locate key data, such as focusing on the data of relevant time periods based on the sleeper's sensitivity to noise, thereby improving the data's pertinence, and facilitating in-depth analysis of the factors affecting sleep quality. Combined with the ambient temperature and the sleeper's physique, it provides an accurate basis for personalized sleep improvement strategies, thereby effectively improving the accuracy and effectiveness of sleep quality analysis.

[0083] Among them, the target sleep data refers to a set of data closely related to sleep quality assessment, sleep pattern analysis and sleep health status judgment that is collected and focused on by the smart bed after comprehensively considering the sleep environment information and the characteristics of the sleeper. It covers information in multiple dimensions, including the sleeper's physiological data during sleep, such as dynamic change curves of heart rate, respiratory rate, and blood oxygen saturation. These data can reflect the sleeper's physical function status and potential respiratory or cardiovascular problems; body movement data, such as the number of turning over, the amplitude and frequency of limb movements, which can be used to infer the sleep depth of the sleeper, whether he is in a state of easy awakening, and the transition of the sleep cycle; it also includes data generated by interaction with the sleep environment, such as fluctuations in sleep duration due to temperature discomfort, the number of sleep interruptions due to noise interference, and the changes in sleep posture and its stability reflected by the mattress pressure distribution data. Optionally, the determination of the target sleep data corresponding to the smart bed can be achieved through a rule-based screening method, such as: determining the target sleep data according to pre-set rules in combination with the sleep environment information and the characteristics of the sleeper.

[0084] The present invention obtains multiple sleep data by segmenting the target sleep data, which can clearly divide the characteristics of different sleep stages, such as distinguishing the data difference between rapid eye movement and non-rapid eye movement periods, which is helpful for in-depth analysis of sleep cycle patterns. On the other hand, it is convenient to accurately locate the time period when sleep problems occur, such as analyzing the data characteristics of insomnia periods, thereby providing strong support for the targeted formulation of sleep improvement plans and improving the accuracy and effectiveness of sleep management.

[0085] Among them, the multiple sleep data refer to several data subsets obtained after segmenting the target sleep data according to the rapid eye movement period and non-rapid eye movement period (including light sleep and deep sleep) in the sleep cycle, for example, the rapid eye movement data segment composed of brain waves, heart rate, breathing rate and other data corresponding to the rapid eye movement period, and the light sleep and deep sleep periods respectively contain independent data sub-segments containing changes in multiple physiological indicators.

[0086] As an embodiment of the present invention, the target sleep data is segmented to obtain multiple sleep data, including: parsing the data timeline corresponding to the target sleep data; determining the sleep cycle interval corresponding to the target sleep data based on the data timeline; analyzing the rapid eye movement period in the sleep cycle interval; identifying the cycle boundary features corresponding to the rapid eye movement period; separating the non-rapid eye movement period in the sleep cycle interval based on the cycle boundary features; and segmenting the target sleep data based on the rapid eye movement period and the non-rapid eye movement period to obtain multiple sleep data.

[0087] Among them, the data timeline refers to the time mark sequence of the complete time period from falling asleep to waking up covered by the target sleep data. It is like a time ruler, providing a measurement benchmark for the order and duration of various data events during the sleep process; the sleep cycle interval refers to the time segment with repetitive and stage characteristics divided on the data timeline according to the physiological laws of human sleep. A complete sleep cycle usually includes a series of regular stage combinations starting from light sleep, gradually transitioning to deep sleep, and then to rapid eye movement. Its length generally fluctuates around 90-120 minutes, and these cycles are arranged in sequence on the timeline; the rapid eye movement period refers to a special stage in the sleep cycle, during which the human eyeball will show rapid and irregular movements, and the brain's neural activity is relatively active, and dreams mostly occur during this period; the cycle boundary feature refers to the iconic data change or data combination feature used to accurately define the start and end positions of the rapid eye movement period. These features can be reflected in multiple physiological data indicators, such as brain wave frequency and vibration. When the brain waves gradually change from the high-amplitude slow waves of deep sleep to the low-amplitude fast waves and are accompanied by rapid fluctuations in eye movement data, it indicates the beginning of the REM period; and when the brain waves again show a trend of changing to high-amplitude slow waves and the eye movements tend to be flat, it marks the end of the REM period; the non-REM period refers to the other stages in the sleep cycle except the REM period, which can be further subdivided into light sleep and deep sleep. The light sleep is the transition stage at the beginning of sleep. The human body's consciousness gradually becomes blurred, but it still has a certain perception of external stimuli. Physiological indicators such as heart rate and breathing begin to slow down, and brain waves show the characteristics of relatively low frequency and high amplitude.

[0088] Furthermore, the analysis of the data time axis corresponding to the target sleep data can be achieved through a timestamp marking method, such as: when collecting target sleep data, an accurate timestamp is assigned to each data point, and the specific time of data collection is recorded at the same time. These data points with timestamps are arranged in sequence to form a data time axis; the determination of the sleep cycle interval corresponding to the target sleep data can be achieved through a power spectrum analysis method, such as: by analyzing the changes in the brain wave power spectrum on the time axis, the time periods with different frequency and power characteristics are determined, thereby dividing the sleep cycle interval; the analysis of the rapid eye movement period in the sleep cycle interval can be achieved through multimodal Data fusion analysis is implemented, such as: combining brain waves, heart rate, breathing rate and other data for comprehensive analysis to obtain the rapid eye movement period; the identification of the periodic boundary characteristics corresponding to the rapid eye movement period can be implemented by a change point detection algorithm, such as: (when the frequency of the brain wave gradually changes from the low frequency and high amplitude of the deep sleep period, the CUSUM algorithm determines it as the beginning boundary of the rapid eye movement period when the cumulative sum of the frequency changes exceeds the set threshold; the segmentation processing of the target sleep data can be implemented by a rule-based segmentation algorithm, such as: after determining the beginning and end boundaries of the rapid eye movement period, the data between the two boundaries are divided into rapid eye movement period data segments.

[0089] The present invention can accurately gain insight into sleep stability by calculating the sleep fluctuation values ​​corresponding to the multiple sleep data. The fluctuation values ​​can be used to intuitively understand the sleep fluctuations of the sleeper in each stage, such as whether the sleeper frequently switches between light sleep and deep sleep, which is helpful to discover potential sleep problems. Larger fluctuation values ​​may indicate sleep disorders and the like, thereby providing a strong basis for targeted improvement of sleep quality and formulation of personalized sleep plans.

[0090] Among them, the sleep fluctuation value refers to a value calculated by a specific formula, which is used to quantify the stability or degree of change of multiple sleep data during the sleep process. It comprehensively considers factors such as the duration of each sleep segment, the average value and standard deviation of physiological indicators in each sleep segment.

[0091] As an embodiment of the present invention, the calculating the sleep fluctuation values ​​corresponding to the multiple sleep data segments includes:

[0092] The sleep fluctuation value corresponding to the multiple sleep data is calculated using the following formula:

[0093]

[0094] Wherein, V represents the sleep fluctuation value corresponding to the multiple sleep data segments, M represents the number of segments into which the sleep data is divided, and t j represents the duration of the jth sleep period, δ j represents the standard deviation of sleep physiological indicators in the jth sleep segment, μ jrepresents the average value of sleep physiological indicators in the jth sleep segment, It represents the overall average value of sleep physiological indicators in all sleep segments.

[0095] Furthermore, the sleep physiological indicators refer to physiological characteristics that can be measured and quantified during the sleep process, such as brain wave frequency, heart rate, breathing rate, eye movement frequency, etc.; the standard deviation refers to the degree of dispersion of the sleep physiological indicator data in the jth sleep segment relative to the average value of the sleep physiological indicators in that segment; the average value refers to the arithmetic mean of all measurement data of the sleep physiological indicators in the jth sleep segment, which represents the typical value or central trend of the sleep physiological indicators in that segment; the overall average value refers to the arithmetic mean of the measurement data of the sleep physiological indicators in all sleep segments, which represents the average level of a physiological indicator during the entire sleep process, and is used to compare with the average value of each sleep segment to measure the degree of deviation of each sleep segment from the overall.

[0096] S3. Identify the number of characteristic indicators corresponding to the multiple sleep data segments, calculate the characteristic deviations corresponding to the multiple sleep data segments based on the characteristic indicators, and perform correction processing on the multiple sleep data segments based on the sleep fluctuation value and the characteristic deviation to obtain sleep correction parameters.

[0097] The present invention can quantify sleep quality by identifying the characteristic index numbers corresponding to the multiple sleep data segments, which is helpful for accurately evaluating sleep quality. The characteristic index numbers can provide a more comprehensive understanding of the conditions of each sleep stage, such as the characteristic manifestations of light sleep and deep sleep, and can provide a quantitative basis for the diagnosis of sleep disorders. When sleep abnormalities are found, specific characteristic index numbers can assist in determining the problem, thereby providing a reliable basis for formulating strategies to improve sleep.

[0098] Among them, the characteristic index number refers to the number of quantitative parameters extracted from multiple sleep data segments and used to characterize the sleep state and characteristics. These parameters may include the proportion of the duration of each sleep stage (such as light sleep, deep sleep, and rapid eye movement), statistical values ​​of physiological indicators such as heart rate, breathing rate, and brain wave frequency during sleep (such as mean value, standard deviation, etc.), and the number of sleep interruptions, turning over frequency, etc. that can reflect the quality and stability of sleep. Optionally, the identification of the characteristic index number corresponding to the multiple sleep data segments can be achieved through a wavelet transform algorithm, such as: performing a wavelet transform on the brain wave data, and the wavelet transform can decompose the brain wave signal into sub-bands of different frequencies, each sub-band corresponding to different sleep stage characteristics.

[0099] The present invention calculates the characteristic deviations corresponding to the multiple sleep data based on the characteristic index number, and can accurately evaluate the sleep quality. The characteristic deviations can be used to understand the difference between the sleep state and the ideal state. Secondly, it helps to discover sleep problems in a timely manner. A larger characteristic deviation indicates potential sleep disorders, thereby providing a strong basis for taking targeted improvement measures.

[0100] The characteristic deviation refers to a value calculated by a specific formula, which is used to measure the degree of deviation of multiple sleep data from the average level in various characteristic indicators.

[0101] As an embodiment of the present invention, the calculating the characteristic deviation corresponding to the plurality of sleep data segments based on the characteristic index number includes:

[0102] The characteristic deviation corresponding to the multiple sleep data segments is calculated using the following formula:

[0103]

[0104] Wherein, D represents the characteristic deviation corresponding to the plurality of sleep data segments, N represents the total number of sleep data segments, i represents the number index of sleep data segments, Q represents the total number of characteristic indicators, p represents the number index of characteristic indicators, and x represents the total number of characteristic indicators. ip represents the actual value of the index corresponding to the pth characteristic index in the i-th sleep data. Represents the average value of the index corresponding to the pth feature index in all sleep data segments.

[0105] Furthermore, the sleep data segment refers to the data of the entire sleep process being divided into multiple parts according to certain rules or methods. For example, the sleep data of one night can be divided into several segments in chronological order, and each segment represents the sleep data for a period of time; the characteristic index refers to various quantitative parameters used to describe the sleep state, which may include the duration ratio of sleep stages (such as light sleep, deep sleep, and rapid eye movement), physiological indicators such as heart rate, breathing rate, and brain wave frequency during sleep, and behavioral indicators such as the number of tossing and turning, and the number of sleep interruptions; the actual value of the indicator refers to the value actually measured of a specific characteristic indicator in a certain section of sleep data, for example, the heart rate value measured in a certain section of sleep data, the actual duration of a certain sleep stage, etc.; the average value of the indicator refers to the average measured value of a characteristic indicator in all sleep data segments, and this average value represents the average level of the characteristic indicator in the entire sleep process.

[0106] The present invention performs correction processing on the multiple sleep data based on the sleep fluctuation value and the characteristic deviation to obtain sleep correction parameters, which can accurately correct the sleep data so that the sleep data more accurately reflects the actual sleep situation; it helps to discover and correct deviations in sleep monitoring, and thus provides a more reliable basis for evaluating sleep quality, diagnosing sleep disorders and formulating personalized sleep improvement plans.

[0107] The sleep correction parameter refers to a parameter further extracted from the weighted data value that can characterize the characteristics of the sleep data after correction. It is the final quantitative indicator used to accurately describe the sleep state after correction based on the sleep fluctuation value and the characteristic deviation.

[0108] As an embodiment of the present invention, the correction processing of the multiple sleep data segments based on the sleep fluctuation value and the characteristic deviation to obtain sleep correction parameters includes: determining the fluctuation deviation coefficient corresponding to each sleep data segment based on the sleep fluctuation value and the characteristic deviation; normalizing the fluctuation deviation coefficient to obtain a normalized deviation value; querying the correction weight corresponding to the normalized deviation value; weighting the multiple sleep data segments according to the correction weight to obtain a weighted data value; and performing parameter extraction on the weighted sum value based on the weighted data value to obtain sleep correction parameters.

[0109] Among them, the fluctuation deviation coefficient refers to a coefficient that comprehensively considers the sleep fluctuation value and the characteristic deviation, and is used to measure the degree of deviation of each sleep data segment relative to the overall sleep stability and characteristic consistency; the normalized deviation value refers to the value obtained after the fluctuation deviation coefficient is processed by a specific normalization method. The purpose of normalization is to map the fluctuation deviation coefficient to a unified, easy-to-compare and easy-to-process numerical range, usually between 0 and 1 or between -1 and 1, etc.; the correction weight refers to the weight value obtained by searching according to the normalized deviation value, which reflects the correction strength for each sleep data segment. Generally speaking, the larger the normalized deviation value, the larger the corresponding correction weight, which means that the sleep data segment deviates further from the ideal state and requires a greater degree of correction; the weighted data value refers to the data value obtained after weighted operation of multiple sleep data segments according to the correction weight, that is, each sleep data segment is multiplied by its corresponding correction weight, and then a comprehensive data value is obtained by summing.

[0110] Furthermore, the determination of the fluctuation deviation coefficient corresponding to each segment of sleep data can be achieved through a principal component analysis algorithm, such as: using the principal component analysis algorithm to convert the original two variables (sleep fluctuation value and characteristic deviation) into a set of new unrelated variables, wherein the first principal component can be used as the fluctuation deviation coefficient; the normalization of the fluctuation deviation coefficient can be achieved through a normalization algorithm, such as: minimum-maximum normalization, Z-score normalization and other algorithms; the query of the correction weight corresponding to the normalized deviation value can be achieved through an interpolation algorithm, such as: linear interpolation and other algorithms; the weighted processing of the multiple segments of sleep data according to the correction weight can be achieved through a weighted summation algorithm, such as: simple weighted summation, hierarchical weighted summation and other algorithms; the parameter extraction of the weighted sum value can be achieved through a statistical feature extraction method, such as: the mean, median, standard deviation and other statistical parameters of the weighted sum value can be extracted as sleep correction parameters.

[0111] S4. Query the key correction indicators in the sleep correction parameters, and perform trend marking on the key correction indicators to obtain trend marking indicators, divide the continuous associated time periods corresponding to the trend marking indicators, and perform trend update on the trend marking indicators based on the continuous associated time periods to obtain trend update indicators.

[0112] By querying the key correction indicators in the sleep correction parameters, the present invention can accurately locate the core problem areas in the sleep data, and can quickly focus on the key factors affecting sleep quality. By clarifying these key indicators, it can efficiently provide a direct basis for personalized sleep improvement plans, which is helpful to adjust sleep habits, environment, etc. in a targeted manner, thereby effectively improving sleep quality and preventing potential sleep disorders.

[0113] Among them, the key correction indicators refer to specific parameters or indicators screened out from the sleep correction parameters, which are of key value for evaluating and improving sleep quality. These indicators can accurately reflect the core characteristics of the sleep state, such as key changes in the proportion of sleep stage duration after correction (such as a significant change in the proportion of deep sleep), the degree of deviation of physiological indicators such as heart rate or respiratory rate from the normal range after correction, key adjustments to behavioral indicators such as the number of sleep interruptions, etc. Optionally, the query of the key correction indicators in the sleep correction parameters can be achieved through indicator query tools, such as: SPSS, Tableau and other tools.

[0114] Furthermore, the present invention obtains trend marking indicators by trend marking the key correction indicators, which can intuitively present the changing trends of key sleep characteristics over time or in different sleep stages, clearly show the dynamic evolution of sleep conditions, help to predict the development trend of sleep quality in advance, and provide accurate guidance for timely adjustment of sleep intervention strategies, thereby more effectively ensuring the continuation and improvement of good sleep quality.

[0115] Among them, the trend marking indicator refers to a new indicator representation formed by appending trend information to the original key correction indicator after analyzing and judging the changing trend of the key correction indicator. For example, for the respiratory rate correction indicator, if its changing trend is a downward trend, then the trend marking indicator can be expressed as "respiratory rate correction value-downward trend".

[0116] As an embodiment of the present invention, the trend marking of the key correction indicator to obtain the trend marking indicator includes: constructing an indicator marking rule corresponding to the key correction indicator; based on the indicator marking rule, collecting the indicator data sequence corresponding to the key correction indicator; extracting the sequence change characteristics corresponding to the indicator data sequence; analyzing the change trend corresponding to the data change characteristics; based on the change trend, trend marking the key correction indicator to obtain the trend marking indicator.

[0117] Among them, the indicator marking rule refers to a set of specifications or criteria formulated based on sleep-related professional knowledge, previous research results and specific analysis purposes. For example, for the sleep stage duration correction indicator, the rule can be set as: if the deep sleep duration correction value increases for three consecutive data points, it is marked as "the initial segment of the rising trend of the deep sleep duration"; if the fluctuation amplitude of the light sleep duration correction value exceeds a certain threshold over a period of time, it is marked as "the light sleep duration high fluctuation segment", etc.; the indicator data sequence refers to a series of values ​​of key correction indicators arranged in a specific order (such as time order, sleep cycle order, etc.). When analyzing the sleep data of an individual for a week, the deep sleep duration correction values ​​in each night's sleep are arranged in sequence. A set of data constitutes a data sequence of the deep sleep duration correction index; the sequence change feature refers to the information that can reflect the inherent change characteristics of the data extracted from the index data sequence through specific mathematical or statistical methods, for example, the first-order difference of the data sequence is calculated to determine the change amount between adjacent data points, so as to judge whether the data is increasing, decreasing or relatively stable; the change trend refers to the overall direction or tendency of the key correction indicator data determined based on the sequence change feature. For example, if the first-order difference of the data sequence of a key correction indicator is mostly positive and the value is relatively stable, and the variance is small, then it can be judged that its change trend is steadily rising; if the first-order difference is alternately positive and negative and the variance is large, it presents an irregular fluctuation trend.

[0118] Furthermore, the construction of indicator labeling rules corresponding to the key correction indicators can be achieved through a cluster analysis algorithm, such as: performing cluster analysis on a large number of samples of key correction indicators in historical sleep data; the collection of indicator data sequences corresponding to the key correction indicators can be achieved through a data screening algorithm, such as: when the sleep monitoring system records a large amount of data including various physiological signals, a data screening algorithm is used to extract the data sequence of key correction indicators; the extraction of sequence change characteristics corresponding to the indicator data sequence can be achieved through feature extraction tools, such as: PCA, LDA and other tools; the analysis of change trends corresponding to the data change characteristics can be achieved through data visualization tools, such as: importing the data sequence of key correction indicators into the Tableau tool, and intuitively analyzing the change trends; the trend marking of the key correction indicators can be achieved through marking tools, such as: Brat, LabelImg and other tools.

[0119] By dividing the continuous associated time periods corresponding to the trend marking indicators, the present invention can clearly show the time stage characteristics of the changes in sleep indicators, so that complex trend changes show stage rules, which is helpful to discover the stability and volatility of sleep states in different time periods, provide a basis for analyzing the dynamic changes in sleep quality, and more accurately formulate targeted sleep improvement strategies and intervention measures.

[0120] Among them, the continuous correlation period refers to a period of time in which the trend marking indicator shows coherent and related changes in the time series. During this period, the trend markings of the key correction indicators related to sleep (such as an upward trend, a downward trend, a stable trend, etc.) have inherent correlation and are not isolated change points. For example, during a continuous period of time, the trend marking of the sleep efficiency correction indicator has always been an upward trend. This complete time range constitutes a continuous correlation period. Optionally, the division of the continuous correlation period corresponding to the trend marking indicator can be achieved by a sliding window method, such as: defining a time window of a fixed size, sliding on the time series of the trend marking indicator, and obtaining a continuous correlation period.

[0121] The present invention updates the trend marking indicator based on the continuous associated time period to obtain a trend update indicator, which can adapt to the dynamic evolution of the sleep state over time, make the trend evaluation more in line with the changes in the actual sleep process, and help to more accurately capture the fluctuation details of sleep quality and timely discover new sleep trend characteristics, thereby providing strong support for formulating more timely and targeted sleep improvement plans, and further improving the analysis and intervention effects on sleep conditions.

[0122] Among them, the trend update index refers to a new index generated by transforming and optimizing the original trend marking index using the trend update framework. For example, after updating the trend marking index of the number of sleep wake-up times, the trend update index can more accurately reflect the real change trend of sleep wake-up conditions in the entire continuous associated period under the influence of those key trend segments (such as periods of abnormal awakening times due to environmental noise or physical discomfort).

[0123] As an embodiment of the present invention, the trend marking indicator is trend updated based on the continuous correlation time period to obtain a trend update indicator, including: determining the initial change feature corresponding to the trend marking indicator within the continuous correlation time period; analyzing the change characteristics corresponding to the initial change feature; based on the change characteristics, screening out key trend segments in the trend marking indicator; based on the key trend segments, constructing a trend update framework corresponding to the trend marking indicator; based on the trend update framework, trend updating the trend marking indicator to obtain a trend update indicator.

[0124] Among them, the initial change feature refers to the specific status information displayed by the trend marking indicator at the starting moment of the continuous association period, covering the initial value of the indicator, for example, in the sleep depth trend marking indicator, what is the initial depth value; it also includes its initial change direction, whether it is an upward trend (such as a gradual transition from light sleep to deep sleep), a downward trend (such as a transition from deep sleep to light sleep) or relatively stable without obvious trend; the change characteristics refer to the essential attributes of the change of the trend marking indicator obtained by further analyzing the initial change characteristics within the continuous association period, including the uniformity of the change, that is, whether the indicator changes at a uniform speed or at a variable speed, for example, the trend marking indicator of the frequency of sleep stage conversion, whether it converts evenly at regular intervals, or the conversion interval is unstable; the key trend segmentation refers to the analysis of the key trend segmentation based on the initial change characteristics. The change characteristics are some sub-periods that are important or representative from the entire change history of the trend marker indicator in the continuous associated time period. For example, in the change process of the trend marker indicator of the number of sleep apnea, if the number of apnea in a certain sub-period not only increases significantly, but also highly overlaps with the rapid eye movement period of the sleep stage, and the change characteristics are similar to the abnormal sleep data of other nights, then this sub-period can be determined as a key trend segment; the trend update framework refers to a set of system architectures built on the key trend segment as the cornerstone for re-evaluating and updating trend marker indicators, which includes quantitative processing rules for key trend segments, such as how to determine the impact weight of the key trend segment on the overall trend marker indicator based on factors such as the change amplitude and duration of the indicator in the key trend segment.

[0125] Furthermore, the determination of the initial change characteristics corresponding to the trend marking indicator within the continuous correlation period can be achieved through Excel tools, such as: using the AVERAGE function in the cell area corresponding to the starting period to calculate the average value to obtain the initial value, and using the STDEV function to calculate the standard deviation to obtain the degree of dispersion information, thereby determining the initial change characteristics; the analysis of the change characteristics corresponding to the initial change characteristics can be achieved through an ARMA model, such as: the ARMA model analyzes the autocorrelation and partial autocorrelation after the initial change characteristics within the continuous correlation period, judges the continuity and periodicity of the change, and thus determines the change characteristics; the screening out of key trend segments in the trend marking indicator can be achieved through screening tools, such as: SQL, DTW, etc.; the construction of the trend update framework corresponding to the trend marking indicator can be achieved through mathematical modeling tools, such as: in the Matlab tool, using its matrix operations and function functions to construct a trend update framework.

[0126] S5. Mining the indicator correlation characteristics between the trend update indicators, constructing a sleep quality analysis strategy corresponding to the smart bed based on the indicator correlation characteristics and the preset life data of the sleeper, performing simulation analysis on the sleep quality analysis strategy to obtain a strategy simulation score, and generating a sleep quality analysis report corresponding to the smart bed based on the strategy simulation score.

[0127] The present invention helps to gain in-depth insights into the intrinsic connections between various aspects of sleep by exploring the indicator correlation characteristics between the trend update indicators, extract systematic rules from complex sleep data, and more accurately grasp the collaborative operation mode of the sleep mechanism, providing a key perspective for a comprehensive and in-depth understanding of the sleep state, thereby providing a strong basis for the formulation and optimization of personalized sleep improvement plans.

[0128] Among them, the indicator correlation characteristics refer to the intrinsic connections and mutual relationship characteristics between trend update indicators, including the synchronization or asynchrony of different trend update indicators in the changing trends, for example, when certain indicators rise, whether other indicators rise synchronously or show opposite changes; the strength of the correlation between indicators reflects the degree of their mutual influence, and strongly correlated indicators tend to have more coordinated performance when the sleep state changes; as well as the causal relationship or accompanying relationship generated by the interaction of indicators, etc. Optionally, the mining of the indicator correlation characteristics between the trend update indicators can be achieved through an association rule mining algorithm, such as: FP-Growth and other algorithms.

[0129] Furthermore, the present invention constructs a sleep quality analysis strategy corresponding to the smart bed based on the indicator correlation characteristics combined with the preset life data of the sleeper, which can comprehensively consider the multiple factors affecting sleep, and integrate the intrinsic connection between sleep indicators with external factors such as daily life and work schedule, so that the analysis results are more in line with the individual's actual sleep situation, which helps to accurately locate the root cause of sleep problems and provide a scientific basis for customizing personalized sleep improvement plans, thereby effectively improving the accuracy and effectiveness of sleep quality assessment and intervention.

[0130] Among them, the preset life data of the sleeper refers to the information about the sleeper's daily living habits and environmental conditions that is determined to be collected at the beginning of building the sleep quality analysis strategy, covering the sleeper's sleeping environment data, such as the noise level, light intensity, and mattress softness and hardness of the bedroom; life rhythm information, such as daily working hours, commuting time, leisure and entertainment time arrangements; and personal habits, including whether there is a habit of reading before going to bed, using electronic devices, whether smoking, etc.; the sleep quality analysis strategy refers to a set of smart beds designed with comprehensive consideration of core quality options for comprehensive and accurate analysis of sleep quality solutions and processes. For example, based on the core quality options, the smart bed focuses on monitoring changes in sleep duration during sleep, analyzing work pressure and pre-sleep relaxation activity data, and establishing association strategies.

[0131] As an embodiment of the present invention, the sleep quality analysis strategy corresponding to the smart bed is constructed based on the indicator association characteristics in combination with the preset life data of the sleeper, including: querying the association influence factor corresponding to the indicator association characteristics; collecting life detail data related to the association influence factor and the preset life data of the sleeper; analyzing the association mapping between the life detail data and the indicator association characteristics; determining the core quality option corresponding to the smart bed based on the association mapping; and constructing the sleep quality analysis strategy corresponding to the smart bed based on the core quality option.

[0132] Among them, the associated influencing factors refer to key elements that are directly or indirectly related to sleep quality and can be linked to life data. For example, if it is found that the frequency of sleep stage transitions is associated with temperature regulation, then factors related to temperature regulation, such as the rate of temperature change and the amplitude of temperature fluctuation, are associated influencing factors; the life detail data refers to more detailed and specific information extracted from the preset life data of the sleeper for the associated influencing factors. For example, when the associated influencing factor is temperature regulation, the life detail data may include the specific temperature value of the sleeper within 1 hour before going to bed, whether warming or cooling equipment is used (such as electric blankets, air conditioning temperature settings), the amount of exercise and sweating before going to bed, etc.; the associated The mapping situation refers to the corresponding relationship pattern of interaction and mutual influence between life detail data and indicator correlation characteristics. For example, analysis shows that when the sleeper uses electronic devices for more than 1 hour before going to bed and the bedroom is brighter, the association between sleep stage transition frequency and body temperature regulation is manifested as increased temperature fluctuations and more frequent and disordered sleep stage transitions; the core quality options refer to the key dimensions or core indicators used to measure sleep quality determined based on the correlation mapping situation. For example, if the correlation mapping shows that sleep duration is closely related to the sleeper's daily work pressure and pre-sleep relaxation activities, then "stress regulation and sleep duration assessment" and "pre-sleep relaxation activities on sleep duration assessment" can become core quality options.

[0133] Furthermore, the query of the associated influencing factors corresponding to the indicator association characteristics can be achieved through a knowledge graph construction tool, such as: constructing a knowledge graph, and through querying and analyzing the knowledge graph, finding out the nodes that are closely connected with the indicator association characteristics as associated influencing factors; the collection of life detail data related to the associated influencing factors and the preset sleeper life data can be achieved through a data mining algorithm, such as: Apriori and other algorithms; the analysis of the association mapping between the life detail data and the indicator association characteristics can be achieved through a Python data analysis library, such as: Pandas, Statsmodels, etc.; the determination of the core quality options corresponding to the smart bed can be achieved through a statistical analysis tool, such as: performing factor analysis in SPSS, inputting a large number of variables related to life detail data and indicator association characteristics, and extracting the main factors through factor analysis. These main factors can be used as core quality options; the construction of a sleep quality analysis strategy corresponding to the smart bed can be achieved through a machine learning platform, such as: TensorFlow and other platforms.

[0134] The present invention simulates and analyzes the sleep quality analysis strategy to obtain a strategy simulation score, which can verify the effectiveness and rationality of the strategy in advance. By simulating different sleep scenarios and individual conditions, potential defects and deficiencies can be discovered, which helps to optimize the accuracy and reliability of the strategy, making the sleep quality assessment more practical, and thus providing smart bed users with a more scientific and personalized sleep quality analysis service, thereby improving the quality and level of overall sleep management.

[0135] Among them, the strategy simulation score refers to the quantitative score given by a specific evaluation algorithm and standard after the sleep quality analysis strategy is tested in a simulated environment according to various preset sleep scenarios and sleeper characteristic data. It comprehensively considers the accuracy of the strategy in judging sleep quality in different simulated scenarios, the completeness of the analysis of key factors, and the degree of fit with the expected results, so as to intuitively reflect the pros and cons and practical application potential of the sleep quality analysis strategy. Optionally, the simulation analysis of the sleep quality analysis strategy can be achieved through simulation analysis methods, such as: Monte Carlo simulation, scenario simulation method and other methods.

[0136] Furthermore, the present invention generates a sleep quality analysis report corresponding to the smart bed based on the strategy simulation scoring, and can present the simulation evaluation results in an intuitive and systematic form, so that the sleep quality status is clear at a glance, which helps to provide users with accurate and personalized sleep feedback and guide users to adjust their living habits and sleeping environment in a targeted manner.

[0137] Among them, the sleep quality analysis report refers to a comprehensive document formed by integrating, analyzing and interpreting the relevant data of the sleeper monitored by the smart bed based on the strategy simulation score. It covers the overall evaluation of the sleep quality of the sleeper in a specific time period, such as sleep depth, sleep duration, sleep continuity and other aspects of quantitative evaluation results; it details the key factors affecting sleep quality, including lifestyle habits (such as pre-sleep activities, work and rest patterns), environmental factors (such as noise, light, temperature) and physiological status (such as heart rate variability, breathing patterns) and other specific manifestations and correlation analysis during sleep; it may also include personalized improvement suggestions based on the evaluation results, such as adjusting work and rest time, optimizing sleep environment settings, etc. Optionally, the generation of the sleep quality analysis report corresponding to the smart bed can be achieved through a report generation tool, such as: JasperReports, FineReport and other tools.

[0138] First of all, by determining the sleep monitoring area corresponding to the smart bed, the present invention can focus on the core range of data collection, avoid data interference from irrelevant areas, and can be more reasonably distributed and configured. While ensuring comprehensive monitoring, it reduces costs and energy consumption, improves the overall performance of the equipment, and thus greatly improves the accuracy and effectiveness of the acquired sleep data. At the same time, based on the sleep environment information and the characteristics of the sleeper, the present invention determines the target sleep data corresponding to the smart bed, and can accurately locate key data. For example, based on the sleeper's sensitivity to noise, it focuses on data for related time periods to improve data pertinence, which helps to deeply analyze the factors affecting sleep quality, and provides an accurate basis for personalized sleep improvement strategies in combination with ambient temperature and the sleeper's physique, thereby effectively improving the accuracy and effectiveness of sleep quality analysis. By identifying the characteristic index numbers corresponding to the multiple sleep data segments, the present invention can quantify sleep quality, which helps to accurately evaluate sleep quality. The characteristic index numbers can provide a more comprehensive understanding of the conditions of each stage of sleep, such as light sleep, deep sleep, and so on. The characteristic performance of sleep can provide a quantitative basis for the diagnosis of sleep disorders. When abnormal sleep is found, the specific characteristic index number can assist in determining the problem, and then provide a reliable basis for formulating strategies to improve sleep. The present invention can accurately locate the core problem area in the sleep data by querying the key correction index in the sleep correction parameter, and can quickly focus on the key factors affecting sleep quality. By clarifying these key indicators, it can efficiently provide a direct basis for personalized sleep improvement plans, which helps to adjust sleep habits, environment, etc. in a targeted manner, thereby effectively improving sleep quality and preventing potential sleep disorders. Further, the present invention helps to deeply understand the intrinsic connection between various elements of sleep by mining the indicator correlation characteristics between the trend update indicators, extract systematic laws from complex sleep data, and more accurately grasp the collaborative operation mode of the sleep mechanism, providing a key perspective for a comprehensive and in-depth understanding of the sleep state, and then providing a strong basis for the formulation and optimization of personalized sleep improvement plans. Therefore, the sleep quality analysis method and system of a smart bed proposed by the present invention can improve the analysis effect of the sleep quality of the smart bed.

[0139] Embodiment 2:

[0140] like Figure 2 , which is a module schematic diagram of a smart bed sleep quality analysis system provided by one embodiment of the present invention.

[0141] The intelligent bed sleep quality analysis system 200 of the present invention can be installed in an electronic device. According to the functions to be implemented, the intelligent bed sleep quality analysis system 200 can include a feature query module 201, a fluctuation value calculation module 202, a calibration processing module 203, a trend update module 204 and a report generation module 205. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0142] In this embodiment, the functions of each module / unit are as follows:

[0143] The feature query module 201 is used to determine the sleep monitoring area corresponding to the smart bed, obtain the sleep environment information corresponding to the sleep monitoring area, and query the sleeper features corresponding to the sleeper in the sleep monitoring area according to the sleep environment information;

[0144] The fluctuation value calculation module 202 is used to determine the target sleep data corresponding to the smart bed based on the sleep environment information and the sleeper characteristics, segment the target sleep data to obtain multiple sleep data, and calculate the sleep fluctuation values ​​corresponding to the multiple sleep data;

[0145] A calibration processing module 203 is used to identify the characteristic index number corresponding to the multiple sleep data, calculate the characteristic deviation corresponding to the multiple sleep data based on the characteristic index number, and perform correction processing on the multiple sleep data based on the sleep fluctuation value and the characteristic deviation to obtain a sleep correction parameter;

[0146] The trend update module 204 is used to query the key correction index in the sleep correction parameter, and trend-mark the key correction index to obtain a trend-marked index, divide the continuous association time period corresponding to the trend-marked index, and trend-update the trend-marked index based on the continuous association time period to obtain a trend update index;

[0147] The report generation module 205 is used to mine the indicator correlation characteristics between the trend update indicators, construct a sleep quality analysis strategy corresponding to the smart bed based on the indicator correlation characteristics and the preset sleeper life data, simulate and analyze the sleep quality analysis strategy to obtain a strategy simulation score, and generate a sleep quality analysis report corresponding to the smart bed based on the strategy simulation score.

[0148] In detail, the modules described in the smart bed sleep quality analysis system 200 described in the embodiment of the present invention adopt the same technical means as the smart bed sleep quality analysis method described in the accompanying drawings when in use, and can produce the same technical effects, which will not be repeated here.

[0149] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for analyzing sleep quality of a smart bed, characterized in that: The method comprises: Determine a sleep monitoring area corresponding to the smart bed, obtain sleep environment information corresponding to the sleep monitoring area, and query sleeper characteristics corresponding to a sleeper in the sleep monitoring area according to the sleep environment information; Based on the sleep environment information and the sleeper characteristics, determining target sleep data corresponding to the smart bed, segmenting the target sleep data to obtain multiple sleep data segments, and calculating sleep fluctuation values ​​corresponding to the multiple sleep data segments; Identify the characteristic index numbers corresponding to the multiple sleep data, calculate the characteristic deviations corresponding to the multiple sleep data based on the characteristic index numbers, and perform correction processing on the multiple sleep data based on the sleep fluctuation values ​​and the characteristic deviations to obtain sleep correction parameters; Querying the key correction index in the sleep correction parameter, and trend-marking the key correction index to obtain a trend-marked index, dividing the continuous association time period corresponding to the trend-marked index, and trend-updating the trend-marked index based on the continuous association time period to obtain a trend update index; The indicator correlation characteristics between the trend update indicators are mined, and a sleep quality analysis strategy corresponding to the smart bed is constructed based on the indicator correlation characteristics and the preset life data of the sleeper. The sleep quality analysis strategy is simulated and analyzed to obtain a strategy simulation score, and a sleep quality analysis report corresponding to the smart bed is generated based on the strategy simulation score.

2. The smart bed sleep quality analysis method according to claim 1, characterized in that: The querying, according to the sleep environment information, the sleeper characteristics corresponding to the sleeper in the sleep monitoring area includes: Determining the information data range in the sleeping environment information; Divide the environmental comfort level corresponding to the information data range; Calculating the environmental tolerance of the sleeper in the sleep monitoring area based on the environmental comfort level; Based on the environmental tolerance, analyzing the environmental reaction tendency of the sleeper in the sleep monitoring area; Based on the environmental reaction tendency, the sleeper characteristics corresponding to the sleeper in the sleep monitoring area are queried.

3. The smart bed sleep quality analysis method according to claim 1, characterized in that: The step of segmenting the target sleep data to obtain multiple segments of sleep data includes: Analyzing the data time axis corresponding to the target sleep data; Based on the data time axis, determining a sleep cycle interval corresponding to the target sleep data; Analyzing the rapid eye movement period in the sleep cycle interval; Identifying period boundary features corresponding to the rapid eye movement period; Separating the non-rapid eye movement period in the sleep cycle interval based on the cycle boundary feature; Based on the rapid eye movement period and the non-rapid eye movement period, the target sleep data is segmented to obtain multiple sleep data.

4. The smart bed sleep quality analysis method according to claim 1, characterized in that: The calculating the sleep fluctuation values ​​corresponding to the plurality of sleep data segments includes: The sleep fluctuation value corresponding to the multiple sleep data is calculated using the following formula: Wherein, V represents the sleep fluctuation value corresponding to the multiple sleep data segments, M represents the number of segments into which the sleep data is divided, and t j represents the duration of the jth sleep period, δ j represents the standard deviation of sleep physiological indicators in the jth sleep segment, μ j represents the average value of sleep physiological indicators in the jth sleep segment, It represents the overall average value of sleep physiological indicators in all sleep segments.

5. The smart bed sleep quality analysis method according to claim 1, characterized in that: The calculating, based on the characteristic index number, characteristic deviations corresponding to the plurality of sleep data segments includes: The characteristic deviation corresponding to the multiple sleep data segments is calculated using the following formula: Wherein, D represents the characteristic deviation corresponding to the plurality of sleep data segments, N represents the total number of sleep data segments, i represents the number index of sleep data segments, Q represents the total number of characteristic indicators, p represents the number index of characteristic indicators, and x represents the total number of characteristic indicators. ip represents the actual value of the index corresponding to the pth characteristic index in the i-th sleep data. Represents the average value of the index corresponding to the pth feature index in all sleep data segments.

6. The intelligent bed sleep quality analysis method according to claim 1, characterized in that: The step of correcting the plurality of sleep data segments based on the sleep fluctuation value and the characteristic deviation to obtain a sleep correction parameter includes: Determining a fluctuation deviation coefficient corresponding to each sleep data segment based on the sleep fluctuation value and the characteristic deviation; Normalizing the fluctuation deviation coefficient to obtain a normalized deviation value; Querying a correction weight corresponding to the normalized deviation value; Performing weighted processing on the plurality of sleep data segments according to the correction weights to obtain weighted data values; Based on the weighted data value, parameter extraction is performed on the weighted sum value to obtain a sleep correction parameter.

7. The intelligent bed sleep quality analysis method according to claim 1, characterized in that: The step of performing trend marking on the key correction indicator to obtain a trend marking indicator includes: Constructing an indicator marking rule corresponding to the key correction indicator; Based on the indicator marking rule, collecting the indicator data sequence corresponding to the key correction indicator; Extracting sequence change characteristics corresponding to the indicator data sequence; Analyze the change trend corresponding to the data change characteristics; Based on the change trend, the key correction indicator is trend-marked to obtain a trend-marked indicator.

8. The intelligent bed sleep quality analysis method according to claim 1, characterized in that: The step of performing trend updating on the trend marking indicator based on the continuous association period to obtain a trend update indicator includes: Determining initial change characteristics corresponding to the trend marking indicator within the continuous correlation period; Analyzing the change characteristics corresponding to the initial change features; Based on the change characteristics, key trend segments in the trend marking indicator are screened out; Based on the key trend segmentation, construct a trend update framework corresponding to the trend marking indicator; Based on the trend update framework, the trend marking indicator is trend updated to obtain a trend update indicator.

9. The intelligent bed sleep quality analysis method according to claim 1, characterized in that: The sleep quality analysis strategy corresponding to the smart bed is constructed based on the indicator correlation characteristics and the preset life data of the sleeper, including: Query the correlation impact factor corresponding to the indicator correlation characteristic; Collecting life detail data related to the associated influencing factors and preset life data of the sleeper; Analyzing the correlation mapping between the life detail data and the indicator correlation characteristics; Based on the association mapping, determining the core quality options corresponding to the smart bed; Based on the core quality options, a sleep quality analysis strategy corresponding to the smart bed is constructed.

10. A smart bed sleep quality analysis system, characterized in that: The system is used to execute the smart bed sleep quality analysis method according to any one of claims 1 to 9, comprising: A feature query module is used to determine the sleep monitoring area corresponding to the smart bed, obtain the sleep environment information corresponding to the sleep monitoring area, and query the sleeper characteristics corresponding to the sleeper in the sleep monitoring area according to the sleep environment information; a fluctuation value calculation module, for determining target sleep data corresponding to the smart bed based on the sleep environment information and the sleeper's characteristics, performing segmentation processing on the target sleep data to obtain multiple segments of sleep data, and calculating sleep fluctuation values ​​corresponding to the multiple segments of sleep data; a calibration processing module, configured to identify characteristic index numbers corresponding to the plurality of sleep data segments, calculate characteristic deviations corresponding to the plurality of sleep data segments based on the characteristic index numbers, and perform correction processing on the plurality of sleep data segments based on the sleep fluctuation values ​​and the characteristic deviations to obtain sleep correction parameters; a trend update module, used for querying the key correction index in the sleep correction parameter, and trend-marking the key correction index to obtain a trend-marked index, dividing the continuous association time period corresponding to the trend-marked index, and trend-updating the trend-marked index based on the continuous association time period to obtain a trend update index; A report generation module is used to mine the indicator correlation characteristics between the trend update indicators, construct a sleep quality analysis strategy corresponding to the smart bed based on the indicator correlation characteristics and preset sleeper life data, simulate and analyze the sleep quality analysis strategy to obtain a strategy simulation score, and generate a sleep quality analysis report corresponding to the smart bed based on the strategy simulation score.

Images