A visual warning device for cardiovascular disease in obese children and method thereof

Abstract

This invention discloses a visual early warning device and method for cardiovascular diseases in obese children, relating to the field of early warning measurement technology. The method includes the following specific steps: S1: Simultaneously collecting cardiovascular-related parameters and status data of children in both clinical and home modes using a flexible wearable device; S2: Correcting the collected physiological parameters through motion artifact filtering and establishing a correlation model between skin conductance response and cardiovascular parameters to correct for emotional interference in diagnosis; S3: Constructing visual interactive interfaces for children and adults to display parameters and provide emotional guidance; S4: Constructing an early warning indicator system based on children's baseline indicators and metabolic-related indicators, classifying early warning levels, and displaying, transmitting, and storing the data through the adult visual interface. This invention improves the convenience and success rate of early warning measurement by determining the emotional state during measurement through motion artifact filtering and emotional interference correction.

Description

Technical Field

[0001] This invention relates to the field of early warning measurement technology, and in particular to a visual early warning device and method for cardiovascular diseases in obese children. Background Technology

[0002] Childhood obesity has become a global public health crisis. According to the "Report on Nutrition and Chronic Diseases of Chinese Residents" and related epidemiological surveys, the overweight and obesity rates among children in my country continue to rise, leading to increasingly prominent complications such as childhood hypertension and early cardiovascular dysfunction. Obesity not only causes abnormal lipid metabolism in children but also triggers changes in cardiac structure, such as left ventricular hypertrophy, and autonomic nervous system dysfunction.

[0003] However, in cardiovascular early warning diagnostic measurements for obese children, traditional cardiovascular function assessments such as electrocardiograms and blood pressure monitoring usually need to be performed in a specific environment within a medical institution. Children are prone to anxiety when facing unfamiliar, large medical equipment, which can lead to abnormal blood pressure and heart rate. Children's anxiety and fear can have an immediate impact on cardiovascular parameters, resulting in false positive measurements. Furthermore, young or active children often find it difficult to remain still and lying flat during the examination, leading to severe motion artifacts, a high data acquisition failure rate, and the need for repeated measurements, which increases the child's resistance.

[0004] Current measurements often only reflect a child's physiological state at a specific resting moment, failing to capture changes in cardiovascular load during real-life scenarios such as sleep and daily activities at home. This is particularly problematic for masked hypertension or nocturnal cardiovascular events. Masked hypertension, where blood pressure is normal in hospital measurements but elevated at home, increases the risk of misdiagnosis and impacts the accuracy of existing measurement and early warning systems. Summary of the Invention

[0005] The purpose of this invention is to provide a visual early warning device and method for cardiovascular diseases in obese children, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a visual early warning method for cardiovascular diseases in obese children, comprising the following specific steps: S1: Simultaneously collect cardiovascular-related parameters and status data of children in both clinical and home settings using flexible wearable devices; S2: The collected physiological parameters are corrected by motion artifact filtering, and a correlation model between skin conductance response and cardiovascular parameters is established to correct for emotional interference in diagnosis. S3: Build visual interactive interfaces for children and adults to display parameters and guide emotions; S4: Construct an early warning indicator system based on children's basic indicators and metabolic-related indicators, classify early warning levels, and display, transmit, and store the data through a visual interface for adults.

[0007] Furthermore, in step S1, the flexible wearable device integrates the following sensors: An electrocardiogram (ECG) sensor, used to collect heart rate variability and ST segment shift parameters; A non-invasive blood pressure sensor for measuring systolic and diastolic blood pressure; A blood oxygen saturation sensor, used to monitor tissue oxygen supply status; A triaxial accelerometer is used to capture the child's movement in real time. A skin conductance sensor, used to identify changes in skin resistance caused by stress in children.

[0008] Furthermore, in step S2, the motion artifact filtering data correction step is as follows: S21: Based on motion state data collected by a triaxial accelerometer, the child's motion state is divided into resting state and active state; S22: Artifact filtering for different motion states: When the exercise state is the resting state, the raw heart rate-related data collected by the electrocardiogram are directly retained; When the exercise state is active, the ECG data is optimized by enhancing the QRS complex to eliminate motion artifacts.

[0009] Furthermore, in step S21, the motion state classification criteria are as follows: Resting state: The movement amplitude collected by the triaxial accelerometer is ≤0.1g and the duration is ≥30s; Activity status: Movement amplitude > 0.1g and duration ≥ 10s.

[0010] Furthermore, in step S22, when the motion state is an active state, the QRS complex frequency in the electrocardiogram is retained by an electronic filter and filtered, and the sliding window integral of the QRS complex is calculated. and dynamic threshold calculation ,when If the duration is ≥8ms, it is considered a valid QRS wave; if the interval between three consecutive peaks is between 150-1000ms, the heart rate calculation is confirmed to be valid; otherwise, it is marked as a motion artifact and removed.

[0011] Furthermore, the sliding window integral equation is:

[0012] Where n is the current discrete-time sampling point of the ECG signal; N = window length; To amplify the square operation of the QRS group amplitude difference; k is a window traversal counting variable used to backtrack to historical data, that is, k starts from 0 and takes values ​​sequentially to traverse all data in the window. When k=0, it is the current point itself s(n-0)=s(n); The equation for calculating the dynamic threshold is:

[0013] in, This is the average value of the real-time window signal. For real-time window standard deviation, coefficient =1.2, =0.8.

[0014] Furthermore, in step S2, the emotional interference correction step is as follows: S23: Collect the mean μ and standard deviation SD of skin conductance response in obese children of the same age and BMI range under resting conditions, and set the emotional stress threshold as μ+2SD; S24: Real-time comparison of the currently collected skin conductance response value with the emotional tension threshold. When the skin conductance response value exceeds the threshold, it is determined that the child is in a state of tension, triggering the emotion regulation process. S25: Blood pressure and heart rate data collected under stress were corrected using age-matched baseline values ​​for obese children in a calm state. These baseline values ​​were then integrated with BMI, adiponectin, and leptin metabolic marker data. A correction model was constructed using a linear regression algorithm, as follows:

[0015] Where k is the correction coefficient obtained by fitting the clinical sample data; To correct the data; This is the original collected data; This is the current skin conductance response value; This is the theoretical baseline value under calm conditions.

[0016] Furthermore, in step S24, the skin conductance response is detected in real time using a flexible wearable device. When the skin conductance response values ​​at multiple consecutive sampling points exceed the threshold, and the measured blood pressure and heart rate values ​​are 10% higher than the calm baseline value, it is determined to be an emotional disturbance state. The emotion regulation process is as follows: Gamified interaction using a child-friendly visual interface helps alleviate emotions. Physiological parameters are converted into rhythms, guiding children to breathe in rhythm. If the skin conductance response value drops below the threshold after emotion regulation, and blood pressure and heart rate return to within ±5% of the individual baseline value, the adjusted measured value is used as valid data. If the skin conductance response value is higher than the threshold after regulation, and blood pressure and heart rate do not return to normal, the corrected baseline value in step S25 is used as valid diagnostic data.

[0017] Furthermore, in step S3, the child's end uses an animated, scenario-based interface to alleviate emotions. When the flexible wearable device detects tension, the adult end uses a visual interface to view and diagnose the emotion.

[0018] A visual early warning device for cardiovascular diseases in obese children, which uses the aforementioned visual early warning method for cardiovascular diseases in obese children.

[0019] The technical effects and advantages of this invention are as follows: (1) This invention breaks through scene limitations and improves measurement convenience by using multimodal flexible wearable devices, avoiding the phenomenon that children cannot be measured due to their active nature. It adopts wristband or chest patch flexible wearable devices to replace traditional rigid medical devices, reducing children's psychological resistance. It integrates ECG, non-invasive blood pressure, blood oxygen, triaxial acceleration and skin conductance response sensors, realizing the synchronous acquisition of multiple parameters in both clinical and home scenarios, solving the problem that children cannot lie down to measure due to their active nature, and improving the convenience and success rate of early warning measurement. (2) This invention uses motion artifact filtering and emotional interference correction settings, combined with the time stamp synchronous analysis of triaxial acceleration data and ECG data, to accurately distinguish between resting and active states, making it easier for parents or doctors to quickly identify the scene in which the data is located, improving the accuracy of early warning data. By setting an emotional tension threshold, the emotional state during measurement is determined, eliminating data abnormalities caused by tension and reducing misjudgments caused by early warning measurements. (3) This invention accurately distinguishes between resting and active states by defining a quantitative standard for the amplitude of movement and a sliding window algorithm. The original data is retained when the state is at rest; when the state is active, the enhanced QRS complex identification algorithm is automatically triggered to eliminate motion artifacts. The data buffering mechanism during state switching effectively avoids data oscillation caused by frequent state fluctuations, ensuring the continuity and accuracy of ECG and blood pressure calculation. (4) This invention introduces a skin conductance sensor as an indicator for emotion assessment, quantifies and corrects emotional interference in pediatric cardiovascular diagnosis, and triggers dual-channel processing when tension is detected: on the one hand, it initiates emotion regulation guidance; on the other hand, it uses a linear regression model combined with baseline values ​​to correct abnormal data, eliminating false increases in blood pressure and heart rate caused by tension, and reducing the missed diagnosis rate and misdiagnosis rate of occult hypertension. (5) This invention constructs a visual interface that separates the child end and the adult end through visualization and gamification interaction, which not only satisfies the child's entertainment psychology, but also meets the professional diagnostic needs of parents and doctors, alleviates measurement anxiety, and constructs a four-level risk warning system based on basic cardiovascular indicators and metabolic related indicators, realizing intuitive risk classification management. Attached Figure Description

[0020] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of the method steps of the present invention; Figure 2 This is a data correction judgment diagram for the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] This invention provides, for example Figures 1-2 This illustrates a visual early warning method for cardiovascular disease in obese children.

[0023] The specific steps include the following: S1: The flexible wearable device can simultaneously collect cardiovascular-related parameters and status data in both clinical and home settings for children. The flexible wearable device is not limited to wristband or chest patch type, which can avoid children directly facing medical equipment and reduce the phenomenon that children cannot lie straight for cardiovascular diagnosis due to their active nature. It also improves the convenience of cardiovascular early warning measurement for obese children. The flexible wearable device integrates an electrocardiogram (ECG) sensor, a non-invasive blood pressure sensor, a blood oxygen saturation sensor, a triaxial accelerometer, and a skin conductance sensor. The ECG sensor is used to collect core cardiovascular parameters such as heart rate variability (HRV) and ST segment deviation. The non-invasive blood pressure sensor measures systolic and diastolic blood pressure, collecting systolic / diastolic blood pressure according to the WS / T610-2019 standard, based on the child's age, gender, and height percentile. The blood oxygen saturation sensor monitors tissue oxygen supply. The triaxial accelerometer captures the child's movement in real time, allowing parents or doctors to quickly distinguish whether the child is at rest or in motion, facilitating the identification of measurement data. The skin conductance sensor identifies changes in skin resistance caused by the child's anxiety, enabling parents or doctors to recognize changes in measurement data caused by emotional interference, reducing misdiagnosis caused by measurement.

[0024] S2: The collected physiological parameters are corrected by motion artifact filtering, and a correlation model between skin conductance response and cardiovascular parameters is established to correct for emotional interference in diagnosis.

[0025] The data collected by the flexible wearable device were used to measure the HRV, ST segment deviation, blood pressure, blood oxygen saturation, exercise status and skin conductance in children. The collected data are used as the measurement physiological parameters for cardiovascular diagnosis in obese children. The data need to be corrected during exercise. The steps for correcting motion artifact filtering data are as follows: S21: Based on motion state data collected by a triaxial accelerometer, the child's motion state is divided into resting and active states. The triaxial accelerometer has a sampling frequency of 50Hz, which is suitable for capturing the child's rapid motion state. The electrocardiogram (ECG) sensor has a sampling frequency of 200Hz, which meets the minimum sampling requirements for QRS complex identification. The data from the triaxial accelerometer and the ECG sensor are synchronized, that is, the acceleration data and ECG data are aligned by timestamp, with a time deviation of ≤10ms. The raw acceleration data is low-pass filtered with a cutoff frequency of 5Hz to remove high-frequency noise such as sensor jitter. The raw ECG data is baseline drift corrected to eliminate baseline shift caused by breathing and body position, laying the foundation for subsequent QRS complex identification. The triaxial accelerometer collects acceleration values ​​along the X, Y, and Z axes, respectively. , , The magnitude of the movement is the absolute value of the resultant acceleration along the three axes, calculated using the following formula:

[0026] The subtraction of 1g is used to eliminate the fundamental influence of gravitational acceleration, retaining only the acceleration changes caused by motion.

[0027] In step S21, the classification criteria for motion states are as follows: Resting state: The movement amplitude collected by the triaxial accelerometer is ≤0.1g and the duration is ≥30s; Activity status: Movement amplitude > 0.1g and duration ≥ 10s; The resting state is determined by a sliding time window, using a 30-second sliding window. The average movement amplitude A of all sampling points within the window is calculated. If A ≤ 0.1g, and the proportion of sampling points with a single movement amplitude exceeding 0.1g within the window is ≤ 5%, and two consecutive sliding windows meet the above conditions, it is determined to be a resting state, and the state start timestamp is marked. The activity status is determined by sliding a window for 10 seconds and calculating the average movement amplitude A within the window. A > 0.1g, and the proportion of sampling points with movement amplitude > 0.1g within the window is ≥ 80% (to avoid misjudgment of accidental actions). If the above conditions are met for one consecutive sliding window, it is determined to be an active state.

[0028] S22: Artifact filtering for different motion states: When the exercise state is the resting state, the raw heart rate-related data collected by the electrocardiogram are directly retained; When the exercise state is active, the ECG data is optimized by enhancing the QRS complex to eliminate motion artifacts; When the motion state changes from resting to active, the ECG data enhancement process is immediately triggered, and the output of raw data is paused. When the motion state changes from active to resting, the enhanced data is continuously output for 30 seconds. After the ECG signal stabilizes, the output of raw data is resumed to avoid data fluctuations caused by frequent state switching.

[0029] In step S22, when the exercise state is active, the QRS complex frequency in the electrocardiogram is retained by an electronic filter and filtered. The sliding window integral of the QRS complex is then calculated. and dynamic threshold calculation ,when If the duration is ≥8ms, it is considered a valid QRS wave; if the interval between three consecutive peaks is between 150-1000ms, the heart rate calculation is confirmed to be valid; otherwise, it is marked as a motion artifact and removed.

[0030] A fourth-order Butterworth bandpass filter (bandpass 5-15Hz) is used to preserve the core frequency of the QRS group and block EMG interference and power frequency interference. The filtering formula is as follows:

[0031] in This is the filtered output electrocardiogram signal. , These represent the filter feedback coefficient and the forward coefficient, respectively; n is the sampling time point of the discrete signal; and k is the filter delay tap count index, i.e., the historical data backtracking offset. The original input ECG signal, This refers to historical ECG signals after a delay. The output value is the recursive delay of the filter itself; for example, the cutoff frequency of 5Hz is... =0.0122, =0.0488, =0.0732, =0.0488, = 0.0122; while the cutoff frequency is 15Hz: = -2.4142, = 2.0358, = -0.8019, = 0.1414; Differential operations, by highlighting the slope characteristics of the QRS group, suppress T-wave and P-wave interference, as shown in the following formula:

[0032] Where n represents the sampling time points of the discrete ECG signal. The output signal from the differential operation represents the slope of the electrocardiogram waveform. To trace back one sampling interval of historical filtered ECG signals; When extracting QRS group features, the amplitude difference between the QRS group and noise is amplified by squaring, as shown in the following formula:

[0033] Where n represents the sampling time points of the discrete ECG signal. This is the feature enhancement signal output after squaring. This is the electrocardiogram slope signal output after differential processing; The sliding window integral equation is:

[0034] Where n is the current discrete-time sampling point of the ECG signal; N = window length, N takes a value of 20, corresponding to a window duration of 100ms; To amplify the square operation of the QRS complex amplitude difference; k is the window traversal counting variable, used to backtrack to historical data, that is, k starts from 0 and takes values ​​sequentially to traverse all data in the window. When k=0, it is the current point itself s(n-0)=s(n). When k=N-1, the farthest and earliest historical data point of the sliding window corresponds to s(n-(N-1)); s(nk) is the ECG characteristic signal after differential flattening processing; The equation for calculating the dynamic threshold is:

[0035] in, This is the average value of the real-time window signal. For real-time window standard deviation, coefficient =1.2, =0.8, when If the duration is ≥8ms, it is considered a valid QRS wave; if the interval between three consecutive peaks is between 150-1000ms (corresponding to a heart rate of 60-400 beats / minute), the heart rate calculation is confirmed to be valid and marked as a valid QRS wave. Artifacts are identified if the interval between candidate peaks is <150ms or >1000ms, or if the amplitude variation coefficient of three consecutive candidate peaks is >30%, it is considered a motion artifact.

[0036] In step S2, a correlation model between skin conductance response and cardiovascular parameters is established to correct for emotional interference. The steps for correcting for emotional interference are as follows: S23: Collect the mean μ and standard deviation SD of skin conductance response in obese children of the same age and BMI range under resting conditions, and set the emotional stress threshold as μ+2SD; S24: Real-time comparison of the currently collected skin conductance response value with the emotional tension threshold. When the skin conductance response value exceeds the threshold, it is determined that the child is in a state of tension, triggering the emotion regulation process. S25: Blood pressure and heart rate data collected under stress were corrected using age-matched baseline values ​​for obese children in a calm state. These baseline values ​​were then integrated with BMI, adiponectin, and leptin metabolic marker data. A correction model was constructed using a linear regression algorithm, as follows:

[0037] Where k is the correction coefficient obtained by fitting the clinical sample data; To correct the data; This is the original collected data; This is the current skin conductance response value; As the theoretical baseline value under calm conditions, the skin conductance response was collected using a constant voltage GSR sensor with a sampling frequency of 1Hz. The sampling site was the inside of the child's wrist, and the sampling site avoided the hair area to reduce interference from hair.

[0038] The corrected blood pressure and heart rate values ​​will replace the original measurements in the subsequent cardiovascular risk assessment process, ensuring that the diagnostic results are not affected by emotional fluctuations. The data in the linear regression model comes from the sample detection collected synchronously in step S1. The skin conductance response sensor, blood pressure, and heart rate sensors use a timestamp synchronization mechanism to ensure that the three physiological parameters are paired and collected within a 500-millisecond time window. When the skin conductance response value is lower than the emotional stress threshold in three consecutive samplings, the system automatically exits the emotion regulation process and resumes the normal monitoring mode. The emotion regulation process includes playing preset soothing audio, adjusting the device interface to warm-toned visual feedback, and providing voice-guided deep breathing instructions, lasting for no less than 90 seconds. Data collection will resume only after the skin conductance response curve shows a downward trend.

[0039] In step S24, the skin conductance response is detected in real time using a flexible wearable device. When the skin conductance response values ​​at multiple consecutive sampling points exceed the threshold, and the measured blood pressure and heart rate values ​​are 10% higher than the calm baseline value, it is determined to be an emotional disturbance state.

[0040] S3: Construct visual interactive interfaces for both children and adults to display parameters and guide emotions. The children's interface is designed using animated scenarios. This animated scenario design is not limited to mapping heart rate parameters to an animated character's running speed, blood pressure parameters to a balloon's inflation state, or SpO2 parameters to a plant's growth state. Children's measured physiological data are directly and in real-time displayed on the adult's visual interactive interface, facilitating doctors' diagnosis and treatment of cardiovascular diseases in obese children.

[0041] In step S3, the child's end uses an animated scene-based interface to alleviate emotions. When the flexible wearable device detects tension, the adult end uses a visual interface to view and diagnose the emotion. The core parameter display area of ​​the adult visual interface uses a color ring chart structure to present the normal / abnormal state of heart rate, blood pressure, and blood oxygen. By setting green as normal, yellow as borderline abnormal, and red as high-risk abnormal, it is easy to intuitively display the child's measurement data. When collecting home measurements, the diagnostic data are used to generate trend analysis charts of weekly / monthly parameter changes. Based on the WS / T610-2019 standard, the standard ranges of percentiles for the same sex, age, and height are superimposed to make it clear to parents that abnormal areas are visible. Furthermore, a heat map is used to present the risk stratification of metabolic subtypes, such as metabolically healthy obesity and metabolically abnormal obesity.

[0042] The emotion regulation process is as follows: Gamified interaction using a visual interface for children can alleviate emotions. Physiological parameters are converted into rhythms, guiding children to breathe in rhythm. If the skin conductance response value drops below the threshold after emotion regulation, and blood pressure and heart rate return to within ±5% of the individual baseline value, the adjusted measured value is used as valid data. If the skin conductance response value is higher than the threshold after regulation, and blood pressure and heart rate do not return to normal, the corrected baseline value in step S25 is used as valid diagnostic data. When the emotion regulation process is triggered, a dynamic breathing waveform animation is displayed to guide children to regulate their breathing by following the waveform rhythm. The GSR change curve is displayed synchronously and in real time on both the child's and adult's devices. An incentive mechanism is also set up so that after a child completes a single effective measurement without significant movement or emotional interference, a virtual badge is unlocked, which reduces children's tension and anxiety when facing medical equipment and reduces the rate of missed diagnosis of occult hypertension.

[0043] S4: Construct an early warning indicator system based on children's basic indicators and metabolic-related indicators, classify early warning levels, and display, transmit, and store the data through a visual interface for adults.

[0044] The basic indicators for early warning are: resting heart rate > age-matched P95, systolic blood pressure / diastolic blood pressure ≥ sex-matched age-height P95, and heart rate variability (SDNNO0ms). Metabolic correlation indicators combine BMI, triglycerides (TG), and the leptin / adiponectin ratio (L / A) to construct a risk score. For example, P = 0.3xBMI + 0.2xTG + 0.5xL / A. The diagnostic risk is determined based on the P value. The normal range for the risk score P is set as P < 1.5, the borderline risk range is 1.5 ≤ P < 2.5, the intermediate risk range is 2.5 ≤ P < 3.5, and the high risk range is P ≥ 3.5.

[0045] The warning levels are divided into four levels: A Level 1 warning indicates the absence of abnormal indicators, corresponding to a risk score of less than 1.5. Level 2 warning is a basic indicator with a critical or risk score of 1.5-2.5. Abnormal indicators are marked with a yellow circular icon on the adult interface, and abnormal periods are highlighted with a yellow dashed line in the trend analysis chart. At the same time, the APP push notification message is triggered: "Please pay attention to the recent fluctuations in children's cardiovascular parameters. It is recommended to monitor regularly and adjust diet and exercise." A Level 3 warning corresponds to two or more basic indicators or metabolic related indicators reaching 2.5-3.5. The interface is marked with an orange circular icon, the abnormal area in the trend analysis chart is filled with orange, and the adult interface is triggered to flash a prompt and push the message "Children have a risk of cardiovascular dysfunction. It is recommended to consult a specialist doctor." A Level 4 warning corresponds to an abnormality in one or more basic indicators (such as resting heart rate > age-matched P99, systolic blood pressure / diastolic blood pressure ≥ sex-matched age-height P99) and a metabolic-related indicator P ≥ 3.5. The interface is marked with a red circular icon, accompanied by vibration alerts on both the child and adult interfaces. Abnormal areas in the trend analysis chart are filled in red and flash. An emergency message is pushed: "Child's cardiovascular condition is high-risk and requires immediate medical evaluation." The complete measurement data and records of exercise or emotional disturbances are also pushed to the relevant medical institution simultaneously.

[0046] Wearable devices connect to mobile terminals (phones / tablets) via Bluetooth 5.0 and transmit data using AES-256 encryption. They support local storage and cloud backup. The cloud database integrates a database of cardiovascular risk factors for childhood obesity, ensuring real-time and secure transmission of physiological parameters between children's and adults' devices with a transmission latency of ≤200ms and a data packet loss rate of <0.5%. They automatically save raw measurement data for the past 3 months and daily / weekly statistical reports. Cloud storage uses a hospital information system (HIS) interface to upload anonymized confirmed case data to the medical cloud platform, facilitating long-term follow-up and multi-center research analysis by doctors.

[0047] A visual early warning device for cardiovascular diseases in obese children, which uses the aforementioned visual early warning method for cardiovascular diseases in obese children.

[0048] Example 1: Cardiovascular diagnosis for childhood obesity was performed through clinical diagnosis. The subjects were obese male children aged 8 years. The baseline information of the children was BMI = 27.3 kg / m2 and height 135 cm. During the measurement, a chest patch-type wearable device was used to simultaneously collect ECG, blood pressure, SpO2, and movement data. The children engaged in mild activity, at which point the accelerometer detected a movement amplitude > 0.5g. The movement status was determined based on the movement data. The results showed a blood pressure measurement of 128 / 82 mmHg, which was considered high according to the P96 standard for children of the same age, sex, and height. The heart rate variability (SDNN) was 85 ms, and the risk score was 3.2, triggering a level 3 warning. The parent's app displayed a red warning icon and pushed a message stating, "The child has a risk of cardiovascular dysfunction. It is recommended to consult a specialist." The doctor could view the complete data through the adult app backend and develop a targeted intervention plan.

[0049] Motion state data collected by a triaxial accelerometer was used to confirm that the dynamic thresholds of the accelerometer in the X, Y, and Z axes were 0.48g, 0.52g, and 0.51g, respectively, all of which met the trigger condition of >0.5g. The algorithm automatically removed the 3.2-second ECG segment affected by motion artifacts, retaining the valid ECG information. For the blood pressure sensor, Kalman filtering was used to control the measurement error within ±3mmHg in children with mild activity, ensuring the clinical reliability of the 128 / 82mmHg reading.

[0050] Heart rate variability analysis was performed within a 5-minute steady-state window. After exercise compensation, the effective RR interval accounted for 94.7%, and the SDNN value of 85ms was lower than the reference mean (112±18ms) for healthy children of the same age, indicating impaired autonomic nervous system regulation. Time-domain analysis showed that the RMSSD was 32ms and the PNN50 was 8.3%. Frequency-domain analysis showed that the low-frequency power (LF) / high-frequency power (HF) ratio increased to 2.4, indicating relatively dominant sympathetic nerve activity. The risk scoring model integrated BMI percentile (P97.2), blood pressure classification (P96), HRV abnormality index, and obesity-related metabolic biomarkers as proxy variables, and the weighted score was 3.2, corresponding to the three-level warning threshold range (3.0-4.0).

[0051] The early warning response mechanism is implemented in layers: the device's local buzzer provides two brief alerts, and the parent's app completes encrypted data upload and cloud analysis within 15 seconds. The red alert interface simultaneously displays a blood pressure trend graph (systolic blood pressure fluctuating between 126-132 mmHg over the past 7 days), a torpedo-shaped HRV scatter plot, and motion interference markers. Push notifications are generated according to the "Guidelines for Cardiovascular Risk Screening in Obese Children in China," including the measurement timestamp, device number, and suggested departments, such as pediatric cardiovascular specialist / pediatric endocrinology. The hospital retrieves the original waveform data through the adult backend. The ECG shows sinus rhythm with occasional premature atrial contractions (incidence 2.1%), and the blood pressure diurnal rhythm analysis lacks the nighttime decline pattern. Considering the high incidence of sleep apnea in obese children, it is recommended to complete 24-hour ambulatory blood pressure monitoring and polysomnography.

[0052] Example 2: This study utilizes home-based monitoring to diagnose cardiovascular disease in obese children aged 12 years. The target population is obese female children with a baseline BMI of 29.1 kg / m2, classified as metabolically abnormal obesity. During the measurement process, a wristband-style device was worn for 24-hour monitoring. At 3 AM, a blood pressure of 125 / 78 mmHg was detected (exceeding the normal nighttime rhythm), which was marked by the system as suspected masked hypertension. After three consecutive days of monitoring, it was confirmed that the nighttime blood pressure remained above P95, triggering a level 4 alert. The adult device displays a monitoring report including blood pressure rhythm curves and emotional disturbance records, and the information is pushed to the associated hospital, where the doctor recommends further echocardiography.

[0053] Echocardiography showed a left ventricular mass index (LVMI) of 38.5 g / m². 7 The patient's systolic blood pressure was at the 97th percentile among children of the same age, suggesting early target organ damage. The interventricular septal thickness was 8.2 mm, and the left ventricular posterior wall thickness was 7.8 mm, both exceeding the upper limit of normal for obese children. Combined with the nighttime average systolic blood pressure load of 42% observed during ambulatory blood pressure monitoring, a diagnosis of masked hypertension complicated by left ventricular aberration was confirmed. Retrieving 72-hour heart rate variability data from the physician's backend showed an elevated low-frequency / high-frequency power ratio (LF / HF) of 2.8, indicating dominant sympathetic nerve activity, forming a chain of evidence linking this to the 4.2 awakenings per hour recorded in fragmented sleep data.

[0054] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A visual early warning method for cardiovascular diseases in obese children, characterized in that, The specific steps include the following: S1: Simultaneously collect cardiovascular-related parameters and status data of children in both clinical and home settings using flexible wearable devices; S2: The collected physiological parameters are corrected by motion artifact filtering, and a correlation model between skin conductance response and cardiovascular parameters is established to correct for emotional interference in diagnosis. S3: Build visual interactive interfaces for children and adults to display parameters and guide emotions; S4: Construct an early warning indicator system based on children's basic indicators and metabolic-related indicators, classify early warning levels, and display, transmit, and store the data through a visual interface for adults.

2. The visual early warning method for cardiovascular diseases in obese children according to claim 1, characterized in that, In step S1, the flexible wearable device integrates the following sensors: An electrocardiogram (ECG) sensor, used to collect heart rate variability and ST segment shift parameters; A non-invasive blood pressure sensor for measuring systolic and diastolic blood pressure; A blood oxygen saturation sensor, used to monitor tissue oxygen supply status; A triaxial accelerometer is used to capture the child's movement in real time. A skin conductance sensor, used to identify changes in skin resistance caused by stress in children.

3. The visual early warning method for cardiovascular diseases in obese children according to claim 1, characterized in that, In step S2, the motion artifact filtering data correction step is as follows: S21: Based on motion state data collected by a triaxial accelerometer, the child's motion state is divided into resting state and active state; S22: Artifact filtering for different motion states: When the exercise state is the resting state, the raw heart rate-related data collected by the electrocardiogram are directly retained; When the exercise state is active, the ECG data is optimized by enhancing the QRS complex to eliminate motion artifacts.

4. The visual early warning method for cardiovascular diseases in obese children according to claim 3, characterized in that, In step S21, the motion state classification criteria are as follows: Resting state: The movement amplitude collected by the triaxial accelerometer is ≤0.1g and the duration is ≥30s; Activity status: Movement amplitude > 0.1g and duration ≥ 10s.

5. The visual early warning method for cardiovascular diseases in obese children according to claim 3, characterized in that, In step S22, when the exercise state is an active state, the QRS complex frequency in the electrocardiogram is retained by an electronic filter and filtered, and the sliding window integral of the QRS complex is calculated. and dynamic threshold calculation ,when If the duration is ≥8ms, it is considered a valid QRS wave; if the interval between three consecutive peaks is between 150-1000ms, the heart rate calculation is confirmed to be valid; otherwise, it is marked as a motion artifact and removed.

6. The visual early warning method for cardiovascular diseases in obese children according to claim 5, characterized in that, The sliding window integral equation is: Where n is the current discrete-time sampling point of the ECG signal; N = window length; To amplify the square operation of the QRS group amplitude difference; k is a window traversal counting variable used to backtrack to historical data, that is, k starts from 0 and takes values ​​sequentially to traverse all data in the window. When k=0, it is the current point itself s(n-0)=s(n); The equation for calculating the dynamic threshold is: in, This is the average value of the real-time window signal. For real-time window standard deviation, coefficient =1.2, =0.

8.

7. The visual early warning method for cardiovascular diseases in obese children according to claim 1, characterized in that, In step S2, the emotional interference correction step is as follows: S23: Collect the mean μ and standard deviation SD of skin conductance response in obese children of the same age and BMI range under resting conditions, and set the emotional stress threshold as μ+2SD; S24: Real-time comparison of the currently collected skin conductance response value with the emotional tension threshold. When the skin conductance response value exceeds the threshold, it is determined that the child is in a state of tension, triggering the emotion regulation process. S25: Blood pressure and heart rate data collected under stress were corrected using age-matched baseline values ​​for obese children in a calm state. These baseline values ​​were then integrated with BMI, adiponectin, and leptin metabolic marker data. A correction model was constructed using a linear regression algorithm, as follows: Where k is the correction coefficient obtained by fitting the clinical sample data; To correct the data; This is the original collected data; This is the current skin conductance response value; This is the theoretical baseline value under calm conditions.

8. The visual early warning method for cardiovascular diseases in obese children according to claim 7, characterized in that, In step S24, the skin conductance response is detected in real time using a flexible wearable device. When the skin conductance response values ​​at multiple consecutive sampling points exceed the threshold, and the measured blood pressure and heart rate values ​​are 10% higher than the resting baseline value, it is determined to be an emotional disturbance state. The emotion regulation process is as follows: Gamified interaction using a child-friendly visual interface helps alleviate emotions. Physiological parameters are converted into rhythms, guiding children to breathe in rhythm. If the skin conductance response value drops below the threshold after emotion regulation, and blood pressure and heart rate return to within ±5% of the individual baseline value, the adjusted measured value is used as valid data. If the skin conductance response value is higher than the threshold after regulation, and blood pressure and heart rate do not return to normal, the corrected baseline value in step S25 is used as valid diagnostic data.

9. The visual early warning method for cardiovascular diseases in obese children according to claim 1, characterized in that, In step S3, the child's end uses an animated, scenario-based interface to alleviate emotions. When the flexible wearable device detects tension, the adult end uses a visual interface to view and diagnose the emotion.

10. A visual early warning device for cardiovascular diseases in obese children, characterized in that, A device for visual early warning of cardiovascular diseases in obese children using the method described in claims 1-9.

Images