Multi-dimensional night urination behavior monitoring system based on rhythm characteristics

Abstract

The invention discloses a multi-dimensional night urination behavior monitoring system based on rhythm features, and relates to the technical field of biomedical signal processing, and the system specifically comprises a data acquisition module, a rhythm feature extraction module, a urine volume dynamics modeling module, a clustering monitoring module and a real-time feedback module; collecting individual data in real time through IoT equipment and preprocessing the individual data; calculating a nocturnal urination frequency index, a nocturnal urination time concentration ratio, a urination interval rhythm variation coefficient and a deviation index based on the individual data, and performing rhythm feature extraction; performing quadratic polynomial least square fitting on the night accumulated urine volume, and calculating a urine volume acceleration index by using an obtained second derivative to quantify a urine volume generation trend; the method comprises the following steps: constructing and preprocessing a night urine multi-dimensional digital phenotypic vector matrix, fitting a Gaussian mixture model based on an expectation maximization algorithm of a Bayesian information criterion, automatically mapping individuals into four types of subtypes according to cluster center features, and outputting individual subtype labels and confidence coefficients; and obtaining a comprehensive risk score through normalized risk assessment.

Description

Technical Field

[0001] This invention relates to the field of biomedical signal processing technology, specifically to a multidimensional nocturia behavior monitoring system based on rhythmic features. Background Technology

[0002] Nocturia is an important symptom in the elderly, patients with chronic diseases, and those in the early stages of neurodegenerative diseases. Traditional monitoring methods rely on subjective questionnaires, voiding diaries, and single-item counts of nocturia. These methods have significant limitations: (1) It cannot characterize the structure of urination rhythm, cannot quantify the diurnal rhythm shift, lacks a calculable “numerical phenotype”, and is difficult to deeply integrate with AI prediction models.

[0003] (2) Although existing wearable devices can monitor sleep, they ignore the fine-grained characteristics of nocturia rhythm, resulting in inaccurate monitoring and inability to achieve real-time intervention and early disease screening.

[0004] (3) Existing nocturia assessment tools only count the number of nocturia, while ignoring the time distribution variance, phase shift and urine volume dynamics trend, resulting in a lack of structure and predictability in monitoring nocturia behavior.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a multidimensional monitoring system for nocturia behavior based on rhythmic features, in order to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a multidimensional nocturia behavior monitoring system based on rhythm features, specifically including: a data acquisition module, a rhythm feature extraction module, a urination dynamics modeling module, a clustering monitoring module, and a real-time feedback module; Data acquisition module: Collects individual data in real time through IoT devices and performs preprocessing; Rhythm feature extraction module: Calculates nocturia frequency index, nocturia time concentration, urination interval rhythm variation coefficient and offset index based on individual data, and extracts rhythm features; Urine dynamics modeling module: By fitting the cumulative nighttime urine volume using the quadratic polynomial least squares method, the obtained second derivative is used to calculate the urine volume acceleration index to quantify the urine volume generation trend, and it is then assembled in series with the rhythm features to construct a multidimensional digital phenotypic vector of nighttime urine. Clustering monitoring module: By constructing and preprocessing the multidimensional digital phenotypic vector matrix of nocturia, the Gaussian mixture model is fitted using the expectation-maximization algorithm optimized based on the Bayesian information criterion, and individuals are automatically mapped into four subtypes based on cluster center features, and the individual subtype labels and confidence scores are output. Real-time feedback module: Obtains comprehensive risk score through normalized risk assessment, uses multi-dimensional indicator cards and phenotypic vector radar chart to realize real-time comparison and display of individual data with the center of the subtype cluster, and establishes a hierarchical alarm mechanism based on preset risk thresholds and indicator fluctuations.

[0008] As a preferred embodiment of the multidimensional nocturia behavior monitoring system based on rhythmic features described in this invention, wherein: The data acquisition module collects the nighttime urination time sequence, urine volume per urination, and sleep time interval in real time through a smart urine meter and a wearable sleep detector; Time series are mapped to a standard nighttime cycle through time normalization; If the time interval between adjacent urinations is less than 30 minutes, it is considered an abnormal urination record and is filtered out.

[0009] As a preferred embodiment of the multidimensional nocturia behavior monitoring system based on rhythmic features described in this invention, wherein: The rhythm feature extraction module counts the total number of nighttime urinations and the number of days in the current cycle, and calculates the nighttime urination frequency index by dividing the total number of nighttime urinations by the number of days in the cycle, which is used to assess the overall severity. The time distribution variance is calculated based on the nocturnal urination time series. The concentration of nocturnal urination time is calculated by taking the reciprocal of the sum of the time distribution variance and the smallest positive number. If nocturnal urination time is concentrated, the time distribution variance is small and the concentration of nocturnal urination time is large, indicating that nocturnal urination events are fixedly distributed throughout the sleep cycle. If nocturnal urination time is diffuse, the time distribution variance is large and the concentration of nocturnal urination time is small, indicating that nocturnal urination events are evenly distributed throughout the sleep cycle. This is used to quantify the degree of clustering of nocturnal urination events. The mean and standard deviation of the intervals are calculated based on a list of adjacent voiding intervals using nocturnal voiding time series. The coefficient of variation of the voiding interval rhythm is calculated by dividing the standard deviation by the mean. If the intervals are stable, the coefficient of variation of the voiding interval rhythm is small, indicating a regular rhythm. If the intervals are random, the coefficient of variation of the voiding interval rhythm is large, indicating a random rhythm. This is used to assess the variability of the voiding interval. By fixing the midpoint of nighttime urination time through nighttime urination time series, the offset index of each nighttime urination time is calculated based on the difference between the nighttime urination time point and the midpoint of the nighttime urination time point. If most nighttime urination occurs in the first half of the night, the offset index of nighttime urination time is negative; if most nighttime urination occurs in the second half of the night, the offset index of nighttime urination time is positive and increases, which is used to quantify the circadian rhythm offset.

[0010] As a preferred embodiment of the multidimensional nocturia behavior monitoring system based on rhythmic features described in this invention, wherein: For each nighttime urination time point, the cumulative urine volume is counted as the target data for fitting and transformed into the data input points for the cumulative fitting curve; The nocturia cumulative function is defined as a quadratic polynomial. The data input points are fitted using the least squares method, and the coefficients of the quadratic polynomial are calculated. Based on the cumulative fitting curve, the overall generation trend of urine volume over time is quantified. The urine volume acceleration index is calculated using the second derivative of the quadratic polynomial. If the coefficient of the quadratic term is greater than 0, then the urine volume acceleration index is greater than 0, indicating that urine volume production is accelerating. If the coefficient of the quadratic term is less than 0, then the urine volume acceleration index is less than 0, indicating that urine production is showing a decelerating trend. If the coefficient of the quadratic term is equal to 0, then the urine volume acceleration index is equal to 0, indicating that urine volume production follows a linear trend. The nocturia multidimensional digital phenotype vector is directly concatenated and assembled according to the rhythm feature extraction order, and the nocturia multidimensional digital phenotype is constructed.

[0011] As a preferred embodiment of the multidimensional nocturia behavior monitoring system based on rhythmic features described in this invention, wherein: The clustering monitoring module constructs a multidimensional digital phenotypic vector matrix of nocturia containing multiple individuals, and performs missing value imputation and standardization processing. Initialize the Gaussian mixture model, setting the number of clusters to 4. Fit the Gaussian mixture model using the expectation-maximization algorithm, including: Each cluster is defined to follow a multivariate Gaussian distribution; Iteratively update the cluster mean, covariance matrix, and mixture weights; Optimize the K value based on the Bayesian information criterion; For each nocturia multidimensional numerical phenotype vector, the posterior probability is calculated and assigned to the cluster k with the highest probability. Subtype label mapping is performed based on the cluster center features, including: Type A (Centered Type): Cluster centers show low coefficient of variation in voiding interval rhythm, low standard deviation of urine volume, and high nocturia frequency index; Type B (diffuse type): Cluster centers show a high coefficient of variation in voiding interval rhythm and a low nocturia frequency index; Type C (second-half-night surge type): Cluster center shows low nocturia frequency index, high urine volume acceleration index, and positive deviation index; Type D (high variability): Cluster centers show high standard deviation of urine volume and high coefficient of variation of voiding interval rhythm; If cluster features do not match the predefined ones, fine-tune the mapping through hierarchical clustering. Output the subtype label and confidence level for each individual.

[0012] As a preferred embodiment of the multidimensional nocturia behavior monitoring system based on rhythmic features described in this invention, wherein: The real-time feedback module extracts subtype labels and corresponding confidence levels based on nightly monitoring, and obtains a comprehensive risk score through normalized risk assessment. The core indicator card displays the nighttime urination frequency, polyuria index, acceleration index, and comprehensive risk score in real time. The multidimensional digital phenotypic vector matrix of nocturia is displayed using a radar chart and compared with the cluster center of the subtype to which the user belongs; Establish tiered alert rules, including: Low-risk alert (overall risk score < 40): Daily morning message: "Nighttime urination was normal last night, continue to maintain good habits"; Medium-risk warning (40≤Comprehensive risk score<70): A notification will be sent within 30 minutes after the nighttime event or the following morning: “Nighttime urination frequency was high last night, it is recommended to reduce evening water intake”; High-risk alert (overall risk score ≥70 or ≥4 times of nocturia for 3 consecutive nights): Instant push notification, and optionally also push notification to the linked family member / doctor: "Increased risk of nocturia, it is recommended to consult a doctor as soon as possible."

[0013] On the other hand, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements the steps of the multidimensional nocturia behavior monitoring system based on rhythm features as described above.

[0014] On the other hand, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements the steps of the multidimensional nocturia behavior monitoring system based on rhythm features as described above in the present invention.

[0015] The technical effects and advantages provided by the present invention in the above technical solution are as follows: (1) In contrast to the traditional simple record of the number of times nocturia occurs, this invention deeply characterizes urination behavior through multidimensional nocturia behavior based on rhythmic features, thereby achieving a highly digitized and refined representation of nocturia symptoms.

[0016] (2) By performing a quadratic polynomial fitting on the cumulative urine volume and using the second derivative to extract the urine volume acceleration index to quantify the dynamic trend (acceleration, deceleration or linearity) of urine volume generation, important dynamic basis is provided for clinical judgment of the cause of nocturnal polyuria.

[0017] (3) Automatic clustering of individuals is achieved using Gaussian mixture model and Bayesian information criterion. By mapping complex phenotypic vectors to four typical subtypes (Type A-D), it is possible to automatically identify populations with significant clinical characteristics such as "centralized" and "diffuse" types, overcoming the subjectivity of manual diagnosis and providing a scientific classification basis for personalized and precise intervention.

[0018] (4) Through the real-time feedback module, the complex mathematical model is transformed into an intuitive radar chart and core indicator cards, enabling horizontal comparison between user data and the center of the subtype cluster. Combined with the graded alarm mechanism (low, medium, and high risk), it can provide immediate behavioral suggestions or medical warnings based on the score, which greatly improves the efficiency of early disease screening and chronic disease management in the home setting. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0020] Figure 1 This is a flowchart of the method for the multidimensional nocturia behavior monitoring system based on rhythm features of the present invention.

[0021] Figure 2 This is a schematic diagram of the modules of the multidimensional nocturia behavior monitoring system based on rhythm features of the present invention. Detailed Implementation

[0022] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0023] Example 1, referring to Figure 1 and Figure 2 This is the first embodiment of the present invention. This embodiment provides a multidimensional nocturia behavior monitoring system based on rhythm features, specifically including: a data acquisition module, a rhythm feature extraction module, a urination dynamics modeling module, a clustering monitoring module, and a real-time feedback module; Data acquisition module: Collects individual data in real time through IoT devices and performs preprocessing; The data acquisition module collects the nighttime urination time sequence, urine volume per urination, and sleep time interval in real time through a smart urine meter and a wearable sleep detector; Time series are mapped to a standard nighttime cycle through time normalization; If the time interval between adjacent urinations is less than 30 minutes, it is considered an abnormal urination record and is filtered out.

[0024] Rhythm feature extraction module: Calculates nocturia frequency index, nocturia time concentration, urination interval rhythm variation coefficient and offset index based on individual data, and extracts rhythm features; The rhythm feature extraction module counts the total number of nighttime urinations and the number of days in the current cycle, and calculates the nighttime urination frequency index by dividing the total number of nighttime urinations by the number of days in the cycle, which is used to assess the overall severity. The time distribution variance is calculated based on the nocturnal urination time series. The concentration of nocturnal urination time is calculated by taking the reciprocal of the sum of the time distribution variance and the smallest positive number. If nocturnal urination time is concentrated (e.g., all in the first half of the night), the time distribution variance is small and the concentration of nocturnal urination time is large, indicating that the distribution of nocturnal urination events is fixed throughout the sleep cycle. If nocturnal urination time is diffuse, the time distribution variance is large and the concentration of nocturnal urination time is small, indicating that the distribution of nocturnal urination events is uniform throughout the sleep cycle. This is used to quantify the degree of clustering of nocturnal urination events. The mean and standard deviation of the intervals are calculated based on a list of adjacent voiding intervals using nocturnal voiding time series. The coefficient of variation of the voiding interval rhythm is calculated by dividing the standard deviation of the intervals by the mean of the intervals. If the intervals are stable (e.g., once every 2 hours), the coefficient of variation of the voiding interval rhythm is small, indicating a regular rhythm. If the intervals are random, the coefficient of variation of the voiding interval rhythm is large, indicating a random rhythm. This is used to assess the variability of the voiding interval. By fixing the midpoint of nighttime urination time through nighttime urination time series, the offset index of each nighttime urination time is calculated based on the difference between the nighttime urination time point and the midpoint of the nighttime urination time point. If most nighttime urination occurs in the first half of the night, the offset index of nighttime urination time is negative; if most nighttime urination occurs in the second half of the night, the offset index of nighttime urination time is positive and increases, which is used to quantify the circadian rhythm offset.

[0025] Urine dynamics modeling module: By fitting the cumulative nighttime urine volume using the quadratic polynomial least squares method, the obtained second derivative is used to calculate the urine volume acceleration index to quantify the urine volume generation trend, and it is then assembled in series with the rhythm features to construct a multidimensional digital phenotypic vector of nighttime urine. For each nighttime urination time point, the cumulative urine volume is counted as the target data for fitting and transformed into the data input points for the cumulative fitting curve; The nocturia cumulative function is defined as a quadratic polynomial. The data input points are fitted using the least squares method, and the coefficients of the quadratic polynomial are calculated. Based on the cumulative fitting curve, the overall generation trend of urine volume over time is quantified. The urine volume acceleration index is calculated using the second derivative of the quadratic polynomial. If the coefficient of the quadratic term is greater than 0, then the urine volume acceleration index is greater than 0, indicating that urine volume production is accelerating. If the coefficient of the quadratic term is less than 0, then the urine volume acceleration index is less than 0, indicating that urine production is showing a decelerating trend. If the coefficient of the quadratic term is equal to 0, then the urine volume acceleration index is equal to 0, indicating that urine volume production follows a linear trend. The nocturia multidimensional digital phenotype vector is directly concatenated and assembled according to the rhythm feature extraction order, and the nocturia multidimensional digital phenotype is constructed.

[0026] Clustering monitoring module: By constructing and preprocessing the multidimensional digital phenotypic vector matrix of nocturia, the Gaussian mixture model is fitted using the expectation-maximization algorithm optimized based on the Bayesian information criterion, and individuals are automatically mapped into four subtypes based on cluster center features, and the individual subtype labels and confidence scores are output. The clustering monitoring module constructs a multidimensional digital phenotypic vector matrix of nocturia containing multiple individuals, and performs missing value imputation and standardization processing. Initialize the Gaussian mixture model, setting the number of clusters to 4. Fit the Gaussian mixture model using the expectation-maximization algorithm, including: Each cluster is defined to follow a multivariate Gaussian distribution; Iteratively update the cluster mean, covariance matrix, and mixture weights; Optimize the K value based on the Bayesian information criterion; For each nocturia multidimensional numerical phenotype vector, the posterior probability is calculated and assigned to the cluster k with the highest probability. Subtype label mapping is performed based on the cluster center features, including: Type A (Centered): Cluster centers show low coefficient of variation of urination interval rhythm, low standard deviation of urine volume, and high nocturia frequency index (if concentrated in a specific time period). Type B (diffuse type): Cluster centers show high coefficient of variation in voiding interval rhythm and low nocturia frequency index (uniform distribution). Type C (second-half-night surge): Cluster center shows low nocturia frequency index, high urine volume acceleration index, and positive deviation index (if integrated rhythm features). Type D (high variability): Cluster centers show high standard deviation of urine volume and high coefficient of variation of voiding interval rhythm; If cluster features do not match the predefined ones, fine-tune the mapping through hierarchical clustering. Output the subtype label and confidence level for each individual.

[0027] Real-time feedback module: Obtains comprehensive risk score through normalized risk assessment, uses multi-dimensional indicator cards and phenotypic vector radar chart to realize real-time comparison and display of individual data with the center of the subtype cluster, and establishes a hierarchical alarm mechanism based on preset risk thresholds and indicator fluctuations. The real-time feedback module extracts subtype labels and corresponding confidence levels based on nightly monitoring, and obtains a comprehensive risk score through normalized risk assessment. The core indicator card displays the nighttime urination frequency, polyuria index, acceleration index, and comprehensive risk score in real time. The multidimensional digital phenotypic vector matrix of nocturia is displayed using a radar chart and compared with the cluster center of the subtype to which the user belongs; Establish tiered alert rules, including: Low-risk alert (overall risk score < 40): Daily morning message: "Nighttime urination was normal last night, continue to maintain good habits"; Medium-risk warning (40≤Comprehensive risk score<70): A notification will be sent within 30 minutes after the nighttime event or the following morning: “Nighttime urination frequency was high last night, it is recommended to reduce evening water intake”; High-risk alert (overall risk score ≥70 or ≥4 times of nocturia for 3 consecutive nights): Instant push notification (delayed to morning in silent mode at night), and optionally also push to the linked family member / doctor: "Increased risk of nocturia, it is recommended to consult a doctor as soon as possible."

[0028] Compared to the traditional method of simply recording the number of times one urinates at night, this invention deeply characterizes urination behavior through multidimensional nocturia behavior based on rhythmic features, achieving a highly digitized and refined representation of nocturia symptoms.

[0029] By performing quadratic polynomial fitting on the cumulative urine volume and using the second derivative to extract the urine volume acceleration index to quantify the dynamic trend (acceleration, deceleration, or linearity) of urine volume generation, important dynamic evidence is provided for clinical judgment of the cause of nocturnal polyuria.

[0030] Automatic clustering of individuals is achieved using Gaussian mixture models and Bayesian information criteria. By mapping complex phenotypic vectors to four typical subtypes (Type A-D), it can automatically identify populations with significant clinical characteristics such as "centralized" and "diffuse" types, overcoming the subjectivity of manual diagnosis and providing a scientific classification basis for personalized and precise intervention.

[0031] Through a real-time feedback module, complex mathematical models are transformed into intuitive radar charts and core indicator cards, enabling horizontal comparison of user data with the center of their respective subtype clusters. Combined with a tiered alert mechanism (low, medium, and high risk), it can provide immediate behavioral suggestions or medical warnings based on the score, greatly improving the efficiency of early disease screening and chronic disease management in home settings. Example 2

[0032] The following is another embodiment of the present invention, which provides a multidimensional nocturia behavior monitoring system based on rhythm characteristics. In order to verify the beneficial effects of the present invention, a simulation experiment is conducted for scientific demonstration.

[0033] This study validates the superiority of our system in multidimensional digital phenotype construction, subtype clustering accuracy, and risk scoring. The key achievement is that, compared to traditional assessment methods that rely solely on the frequency of nocturia, our system can accurately identify hidden abnormalities (such as surge nocturia in the second half of the night), thus achieving higher clinical early warning value.

[0034] Patients representing the four subtypes were selected, and seven nighttime data were simulated for each case, with the nighttime cycle normalized to [0, 1]. Using Python to simulate IoT devices, real-time data collection was performed on nighttime urination time series, urine volume per urination, and sleep time intervals. The actual calculation results after executing the rhythm feature extraction module are shown in the table below: Subtype Nocturia frequency index Nocturia time concentration Coefficient of variation of urination interval rhythm Nocturia deviation index Average urine volume Standard deviation of urine volume Urine volume acceleration index Nocturia multidimensional numerical phenotype vector matrix Type A (Centralized) 0.571 13.635 0.102 -0.295 200.0 9.1 -757.6 [0.571,13.635,0.102,-0.295,200.0,9.1,-757.6] Type B (Diffuse type) 0.571 3.858 0.000 -0.050 162.5 6.5 -62.5 [0.571,3.858,0.000,-0.050,162.5,6.5,-62.5] Type C (Late Night Surge Type) 0.571 4.996 0.136 0.190 217.5 74.1 4108.3 [0.571,4.996,0.136,0.190,217.5,74.1,4108.3] Type D (Highly variable) 0.571 2.870 0.610 -0.045 210.0 125.2 687.1 [0.571,2.870,0.610,-0.045,210.0,125.2,687.1] Table 1 The urine volume acceleration index was used to construct a 12×7 DVP matrix (3 noisy samples per subtype), and the Gaussian mixture model was fitted using the expectation-maximization algorithm based on Bayesian information criterion optimization. Representative samples were all correctly mapped to their corresponding subtypes (Type A: low CV + high NVDI; Type B: low NFI + uniform distribution; Type C: positive PSI + high UAI; Type D: high σ_v + high CV), with confidence levels all > 95%; Calculate the overall risk score R (0-100 points) using Type C: The system's risk score is 100.0 (high risk, driven by UAI=4108.3 and PSI=0.190). Traditional method (NFI only) risk score = 11.4 (low risk, completely ignoring the late-night surge).

[0035] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A multidimensional nocturia behavior monitoring system based on rhythmic features, characterized in that, Specifically, it includes: The system includes a data acquisition module, a rhythm feature extraction module, a urinary kinetic modeling module, a cluster monitoring module, and a real-time feedback module. Data acquisition module: Collects individual data in real time through IoT devices and performs preprocessing; Rhythm feature extraction module: Calculates nocturia frequency index, nocturia time concentration, urination interval rhythm variation coefficient and offset index based on individual data, and extracts rhythm features; Urine kinetic modeling module: By fitting the cumulative nighttime urine volume using the quadratic polynomial least squares method, the obtained second derivative is used to calculate the urine volume acceleration index to quantify the urine volume generation trend, and it is then connected and assembled with the rhythm features to construct a multidimensional digital phenotypic vector of nighttime urine. Clustering monitoring module: By constructing and preprocessing the multidimensional digital phenotype vector matrix of nocturia, the Gaussian mixture model is fitted using the expectation-maximization algorithm optimized based on the Bayesian information criterion, and individuals are automatically mapped into four subtypes based on the cluster center features, and the individual subtype labels and confidence scores are output. Real-time feedback module: Obtains comprehensive risk score through normalized risk assessment, uses multi-dimensional indicator cards and phenotypic vector radar chart to realize real-time comparison and display of individual data with the center of the subtype cluster, and establishes a hierarchical alarm mechanism based on preset risk thresholds and indicator fluctuations.

2. The multidimensional nocturia behavior monitoring system based on rhythmic features according to claim 1, characterized in that: The data acquisition module collects the nighttime urination time sequence, urine volume per urination, and sleep time interval in real time through a smart urine meter and a wearable sleep detector. Time series are mapped to a standard nighttime cycle through time normalization; If the time interval between adjacent urinations is less than 30 minutes, it is considered an abnormal urination record and is filtered out.

3. The multidimensional nocturia behavior monitoring system based on rhythmic features according to claim 1, characterized in that: The rhythm feature extraction module counts the total number of nighttime urinations and the number of days in the current cycle, and calculates the nighttime urination frequency index by dividing the total number of nighttime urinations by the number of days in the cycle. The time distribution variance is calculated based on the nocturnal urination time series, and the nocturnal urination time concentration is calculated by the reciprocal of the sum of the time distribution variance and the small positive number. The interval mean and interval standard deviation were calculated based on a list of adjacent voiding intervals using nocturnal voiding time series. The interval standard deviation was divided by the interval mean to calculate the urination interval rhythm variation coefficient. The midpoint of nighttime urination time is fixed by using the nighttime urination time series, and the offset index of each nighttime urination time is calculated by subtracting the midpoint of the nighttime urination time from the nighttime urination time point.

4. The multidimensional nocturia behavior monitoring system based on rhythmic features according to claim 1, characterized in that: For each nighttime urination time point, the cumulative urine volume is counted as the target data for fitting and transformed into the data input points for the cumulative fitting curve; The nocturia cumulative function is defined as a quadratic polynomial. The data input points are fitted using the least squares method, and the coefficients of the quadratic polynomial are calculated. Based on the cumulative fitting curve, the overall generation trend of urine volume over time is quantified. The urine volume acceleration index is calculated using the second derivative of the quadratic polynomial. If the coefficient of the quadratic term is greater than 0, then the urine volume acceleration index is greater than 0, indicating that urine volume production is accelerating. If the coefficient of the quadratic term is less than 0, then the urine volume acceleration index is less than 0, indicating that urine production is showing a decelerating trend. If the coefficient of the quadratic term is equal to 0, then the urine volume acceleration index is equal to 0, indicating that urine volume production follows a linear trend. The nocturia multidimensional digital phenotype vector is directly concatenated and assembled according to the rhythm feature extraction order, and the nocturia multidimensional digital phenotype is constructed.

5. The multidimensional nocturia behavior monitoring system based on rhythmic features according to claim 1, characterized in that: The clustering monitoring module constructs a multidimensional digital phenotypic vector matrix of nocturia containing multiple individuals, and performs missing value imputation and standardization processing. Initialize the Gaussian mixture model, setting the number of clusters to 4. Fit the Gaussian mixture model using the expectation-maximization algorithm, including: Each cluster is defined to follow a multivariate Gaussian distribution; Iteratively update the cluster mean, covariance matrix, and mixture weights; Optimize the K value based on the Bayesian information criterion; For each nocturia multidimensional numerical phenotype vector, the posterior probability is calculated and assigned to the cluster k with the highest probability. Subtype label mapping is performed based on the cluster center features, including: Type A: Cluster centers show low coefficient of variation in voiding interval rhythm, low standard deviation of urine volume, and high nocturia frequency index; Type B: Cluster centers show a high coefficient of variation in voiding interval rhythm and a low nocturia frequency index; Type C: Cluster center displays low nocturia frequency index, high urine volume acceleration index, and positive deviation index; Type D: Cluster centers show high urine volume standard deviation and high coefficient of variation of voiding interval rhythm; If cluster features do not match the predefined ones, fine-tune the mapping through hierarchical clustering. Output the subtype label and confidence level for each individual.

6. The multidimensional nocturia behavior monitoring system based on rhythmic features according to claim 1, characterized in that: The real-time feedback module extracts subtype labels and corresponding confidence levels based on nightly monitoring, and obtains a comprehensive risk score through normalized risk assessment. The core indicator card displays the nighttime urination frequency, polyuria index, acceleration index, and comprehensive risk score in real time. The multidimensional digital phenotypic vector matrix of nocturia is displayed using radar charts and compared with the cluster centers of the individual's subtype. Establish tiered alert rules, including: Low-risk reminder: Daily morning message: "Nighttime urination was normal last night, continue to maintain good habits"; Medium-risk warning: A notification will be sent within 30 minutes of the nighttime event or the following morning stating, "You urinated more frequently last night; it is recommended to reduce your water intake in the evening." High-risk alert: Instant push notification, and optionally also push notification to linked family members / doctors: "Increased risk of nocturia, it is recommended to consult a doctor as soon as possible." 7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements a module of the multidimensional nocturia behavior monitoring system based on rhythm features as described in any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements a module of the multidimensional nocturia behavior monitoring system based on rhythm features as described in any one of claims 1 to 6.

Images