Pet heart rate monitoring system based on cloud computing

Abstract

The invention relates to the technical field of remote biomedical monitoring, in particular to a pet heart rate monitoring system based on cloud computing, which comprises a heart rate fluctuation recording module, a motion mode tracking module, an abnormal marking module, a grading evaluation module and an abnormal signal triggering module. According to the method, real-time mammalian pet heart rate data are collected and analyzed, so that the real-time performance and accuracy of pet health monitoring are remarkably improved, heart rate fluctuation is captured in real time, changes in a short time are recognized, the monitoring system can quickly respond to pet health changes, potential health problems are recognized in time, and the pet health monitoring efficiency is improved. The synchronous collection of data is accelerated to further enrich the recognition of motion modes, the change of stride frequency and motion direction is analyzed, more comprehensive understanding of pet motion modes is provided, the heart rate and motion data are synthesized, the recognition of abnormal states is more accurate, the correlation with motion states can be rapidly evaluated when the heart rate is abnormal, and the accuracy of the abnormal heart rate is improved. And the pet health risk can be predicted and identified more effectively.

Description

Technical Field

[0001] The present invention relates to the technical field of remote biomedical monitoring, and particularly to a pet heart rate monitoring system based on cloud computing. Background Art

[0002] The technical field of remote biomedical monitoring involves using information technology and biomedical engineering technology to remotely track and evaluate the health status of patients or organisms. The core content covers data collection, transmission, storage, and analysis. This technical field mainly focuses on collecting vital sign data, such as heart rate, blood pressure, or body temperature, through sensors, and transmitting the data to a remote server or cloud platform via wireless communication technology. The cloud platform is responsible for storing and analyzing the data for remote access by medical professionals or users.

[0003] Among them, a pet heart rate monitoring system based on cloud computing refers to a heart rate monitoring solution specifically designed for mammalian pets. Its core is to collect heart rate data through sensors worn on mammalian pets. The data is sent to a cloud-based storage and processing system. The technical matters targeted by the patent theme include real-time collection of data and rapid transmission to the cloud platform. The system manages and analyzes the collected data through cloud computing to ensure the accuracy and real-time nature of the data. In this way, pet owners can instantly obtain the health status of their pets without complex equipment or in-depth technical knowledge.

[0004] The existing technologies mainly rely on intermittent data collection and analysis, and cannot achieve real-time synchronous analysis of heart rate and behavior data, resulting in data processing delays and missing key health anomaly signals. Most existing systems focus on single vital sign monitoring, such as separate heart rate or step frequency analysis, and lack the ability to comprehensively process multiple data sources. This limitation makes it difficult for the system to instantly and accurately capture changes in a dynamic environment when a pet has rapid heart rate changes or behavior pattern changes, affecting the comprehensive assessment of the health status. The lack of real-time data analysis function also reduces the probability of preventive intervention, resulting in pet owners not receiving timely alerts when their pets' health problems occur. Summary of the Invention

[0005] The purpose of the present invention is to solve the drawbacks existing in the prior art, and to propose a pet heart rate monitoring system based on cloud computing.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions. A pet heart rate monitoring system based on cloud computing includes:

[0007] A heart rate fluctuation recording module collects heart rate data of mammalian pets, identifies the amplitude of heart rate changes within a short period of time, calculates the heart rate change rate, analyzes consecutive multiple heart rate fluctuation segments, and generates heart rate fluctuation characteristics;

[0008] The motion pattern tracking module uses the heart rate fluctuation characteristics to collect the acceleration data of the pet, analyzes the change of the step frequency and the motion direction, identifies the stability of the gait during the motion, and generates a motion pattern trajectory;

[0009] The abnormal marking module compares the heart rate fluctuation of the pet under different motion states through the motion pattern trajectory, identifies the corresponding relationship between the heart rate and the motion pattern, analyzes the heart rate change range during motion and at rest, records the situation where the heart rate fluctuation exceeds the normal range, marks the abnormal state, and generates an abnormal state recognition result;

[0010] The grading evaluation module uses the abnormal state recognition result to analyze the continuous motion duration and interruption of the pet when the heart rate abnormality occurs, divides different levels, and obtains a heart rate abnormality grading result;

[0011] The abnormal signal triggering module determines whether the pet's heart rate abnormality exceeds the triggering threshold according to the heart rate abnormality grading result. If it exceeds, a remote abnormal warning signal is pushed to obtain a heart rate monitoring abnormal warning result.

[0012] As a further solution of the present invention, the heart rate fluctuation characteristics include the peak value, valley value, average heart rate, and heart rate fluctuation frequency of the heart rate. The motion pattern trajectory includes step frequency statistics, motion trajectory mapping, and gait stability analysis results. The abnormal state recognition result includes the time stamp of the heart rate abnormality, the abnormal duration, and the abnormal heart rate interval. The heart rate abnormality grading result is the analysis result of the heart rate fluctuation trend, the abnormal severity rating, and the potential factors affecting the motion. The heart rate monitoring abnormal warning result includes the alarm triggering threshold setting, the alarm sending time, and the alarm status update information.

[0013] As a further solution of the present invention, the acquisition steps of the heart rate fluctuation characteristics are specifically as follows:

[0014] The heart rate signal acquisition sub-module collects the heart rate data of the pet, detects the continuous heart rate of the pet, screens the complete heart rate segments, deletes the abnormal data, and generates a heart rate data sequence;

[0015] The heart rate change amplitude recognition sub-module, based on the heart rate data sequence, identifies the heart rate change amplitude and time interval of the pet within a continuous time segment, and uses the formula:

[0016]

[0017] Calculates the heart rate change rate value, and obtains the heart rate fluctuation rate trend according to the time distribution;

[0018] wherein, R v represents the heart rate change rate value, H i represents the heart rate value at the i-th time point, Hi+1 represents the heart rate value at the (i + 1)-th time point, T i represents the corresponding time value at the i-th time point, T i+1 represents the corresponding time value at the (i + 1)-th time point, and n is the number of time points;

[0019] The heart rate fluctuation stability evaluation sub-module uses the trend of the heart rate fluctuation rate to identify continuous heart rate fluctuation segments, identify the rate differences between adjacent segments, determine abnormal fluctuation points, and generate heart rate fluctuation characteristics.

[0020] As a further solution of the present invention, the steps for obtaining the motion mode trajectory are specifically as follows:

[0021] The step frequency recognition sub-module synchronously collects the acceleration data during the pet's movement based on the heart rate fluctuation characteristics, identifies the changes in acceleration within different time intervals, and uses the formula:

[0022]

[0023] Calculate the step frequency and the direction change value, and perform zonal judgment based on the step frequency and the direction change value to obtain the step frequency and the change in the movement direction;

[0024] where F bf represents the step frequency and the direction change value, and respectively represent the accelerations in the x-axis and y-axis directions at different time points, A is the number of acceleration measurements within the time interval, v b represents the speed within the different time intervals, θ b is the movement direction angle corresponding to the speed, B is the number of speed measurements, H c is the heart rate at different time points, is the average heart rate, and C is the number of heart rate measurements;

[0025] The gait stability analysis sub-module identifies the step frequency fluctuation interval and the trend of the movement direction change through the step frequency and the change in the movement direction, identifies the gait time stability interval under different movement states, and generates the motion mode trajectory.

[0026] As a further solution of the present invention, the steps for obtaining the abnormal state recognition result are specifically as follows:

[0027] The motion mode comparison sub-module uses the motion mode trajectory to extract the pet's heart rate data under different movement states, compares the heart rate fluctuations under different movement states, calculates the difference between the heart rate fluctuation value under the movement state and the heart rate reference value in the static state, and generates the heart rate fluctuation difference value;

[0028] The heart rate variation range analysis sub-module analyzes the variation of the heart rate range in the exercise state and the rest state based on the heart rate fluctuation difference value, in combination with the heart rate data in the exercise state and the heart rate data in the rest state, using the formula:

[0029]

[0030] Calculate the difference value of the heart rate variation range to obtain the heart rate fluctuation range;

[0031] Among them, H d represents the difference value of the heart rate variation range, H mf represents the heart rate value in the f-th exercise state, H sf represents the heart rate value in the f-th rest state, represents the average value of the heart rate values in the exercise state, min(H m ) represents the minimum heart rate in the exercise state, max(H s ) represents the maximum heart rate in the rest state, and F represents the total number of sample data;

[0032] The fluctuation situation identification sub-module uses the heart rate fluctuation range to compare the real-time heart rate fluctuation with the set normal heart rate fluctuation threshold, judges the heart rate fluctuation situation exceeding the threshold, and generates an abnormal state identification result.

[0033] As a further solution of the present invention, the specific steps for obtaining the heart rate abnormality grading result are as follows:

[0034] The exercise state analysis sub-module uses the abnormal state identification result to record the exercise time and the interval time before and after the abnormal heart rate of the pet, extracts the exercise state change trend, judges whether the abnormal heart rate is caused by exercise, calculates the exercise duration ratio corresponding to the abnormal heart rate, using the formula:

[0035]

[0036] Calculate to obtain the exercise state abnormal correlation rate;

[0037] Among them, R m represents the exercise state abnormal correlation rate, T mg represents the g-th exercise duration before the abnormal heart rate, A mg represents the g-th exercise intensity before the abnormal heart rate, T bg represents the g-th exercise interval time after the abnormal heart rate, and G represents the total amount of exercise data corresponding to the abnormal heart rate;

[0038] The heart rate abnormality grading sub-module, based on the exercise state abnormal correlation rate, identifies the heart rate fluctuation trend corresponding to different abnormal types, counts the abnormal fluctuation frequency and duration, divides the heart rate abnormality level, and obtains the heart rate abnormality grading result.

[0039] As a further solution of the present invention, the step of obtaining the abnormal warning result of heart rate monitoring is specifically as follows:

[0040] The abnormal signal recognition sub-module uses the heart rate abnormal grading result to obtain the heart rate data corresponding to the abnormal state, combines the heart rate deviation amplitude, fluctuation frequency and duration, and adopts the formula:

[0041]

[0042] Calculate the abnormal signal strength, compare with the trigger threshold, screen the abnormal states that meet the over-limit standard, and generate the abnormal signal strength screening result;

[0043] Among them, AS represents the abnormal signal strength, H represents the real-time heart rate value, H ref represents the reference heart rate value, D o represents the heart rate fluctuation deviation value, Q represents the number of heart rate samples within the calculation window, W represents the weight parameter, T var represents the heart rate time variability;

[0044] The remote warning push sub-module uses the abnormal signal strength screening result to judge whether the remote warning condition is met. If it is met, a remote abnormal warning signal is pushed, and the abnormal state is recorded to generate the abnormal warning result of heart rate monitoring.

[0045] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0046] In the present invention, by collecting and analyzing the real-time heart rate data of mammalian pets, the real-time performance and accuracy of pet health monitoring are significantly improved, the heart rate fluctuations are captured in real time, and the changes within a short time are identified, so that the monitoring system can quickly respond to the pet health changes and timely identify potential health problems. Accelerating the synchronous collection of data further enriches the recognition of exercise patterns, and by analyzing the changes in stride frequency and movement direction, a more comprehensive understanding of the activity patterns of mammalian pets is provided. Combining heart rate and exercise data makes the recognition of abnormal states more accurate, and can quickly evaluate the correlation with the exercise state when the heart rate is abnormal, and more effectively predict and identify the health risks of mammalian pets. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is the system flow chart of the present invention;

[0048] Figure 2 is the flow chart of the heart rate fluctuation recording module of the present invention;

[0049] Figure 3 is the flow chart of the exercise pattern tracking module of the present invention;

[0050] Figure 4Flow chart of the abnormal marking module of the present invention;

[0051] Figure 5 Flow chart of the hierarchical evaluation module of the present invention;

[0052] Figure 6 Flow chart of the abnormal signal triggering module of the present invention. Detailed implementation manners

[0053] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. 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.

[0054] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by terms such as "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more unless otherwise specifically defined.

[0055] Please refer to Figure 1 , a pet heart rate monitoring system based on cloud computing includes:

[0056] The heart rate fluctuation recording module collects the heart rate data of mammalian pets, identifies the amplitude of heart rate changes within a short period of time, calculates the heart rate change rate, analyzes consecutive multiple heart rate fluctuation segments, evaluates the stability of heart rate fluctuations, identifies abnormal heart rate fluctuation points, and generates heart rate fluctuation characteristics;

[0057] The motion mode tracking module uses the heart rate fluctuation characteristics, collects the acceleration data of the pet, analyzes the change of step frequency and motion direction, identifies the stability of the gait during the motion, extracts the characteristics under different motion states, and generates a motion mode trajectory;

[0058] The abnormal marking module compares the heart rate fluctuations of the pet under different motion states through the motion mode trajectory, identifies the corresponding relationship between the heart rate and the motion mode, analyzes the heart rate change intervals during motion and at rest, records the situation where the heart rate fluctuation exceeds the normal range, marks the abnormal state, and generates an abnormal state recognition result;

[0059] The hierarchical evaluation module uses the abnormal state recognition result, analyzes the duration and interruption of the pet's motion when the heart rate abnormality occurs, combines the fluctuation trend of the heart rate abnormality, divides different levels, and obtains a heart rate abnormality grading result;

[0060] Based on the heart rate abnormality grading result, the abnormal signal trigger module determines whether the pet's heart rate abnormality exceeds the trigger threshold. If it exceeds, a remote abnormal warning signal is pushed, and the continuous monitoring status is recorded to obtain the heart rate monitoring abnormal warning result;

[0061] The heart rate fluctuation characteristics include the peak value, valley value, average heart rate, and heart rate fluctuation frequency of the heart rate. The motion mode trajectory includes step frequency statistics, motion trajectory mapping, and gait stability analysis results. The abnormal state recognition result includes the time stamp of the heart rate abnormality, the abnormal duration, and the abnormal heart rate interval. The heart rate abnormality grading result is the division of the heart rate fluctuation trend, the evaluation of the abnormal severity, and the analysis result of the potential factors affecting the motion. The heart rate monitoring abnormal warning result includes the alarm trigger threshold setting, the time when the alarm is issued, and the alarm status update information.

[0062] Please refer to Figure 2 , and the specific steps for obtaining the heart rate fluctuation characteristics are as follows:

[0063] The heart rate signal acquisition sub-module collects the pet's heart rate data, detects the pet's continuous heart rate, screens complete heart rate segments, deletes abnormal data, and generates a heart rate data sequence;

[0064] It is necessary to obtain the pet's heart rate data within a set time period through a heart rate sensing device. Taking an adult dog as an example, if the monitoring time is set to 24 hours and the heart rate value is recorded once per minute, 1440 heart rate data points can be collected. Continuity detection is performed on the collected data. By identifying the time stamp sequence, it is judged whether the time interval between two adjacent pieces of data is one minute. If the interval exceeds one minute, this part of the data is marked as missing data and excluded. If the interval is two minutes between the 100th minute and the 101st minute, the data at this point is determined to be abnormal. The heart rate segments are screened, and the screening criteria are that the continuous heart rate recording length is not less than 60 minutes, and the heart rate fluctuation per minute does not exceed 20 bpm. If in the selected segment, the heart rate suddenly jumps from 80 bpm to 120 bpm in a certain minute, it is regarded as an abnormal fluctuation and the segment is excluded. The selected continuous heart rate data is sorted to form a heart rate data sequence.

[0065] Based on the heart rate data sequence, the heart rate change amplitude recognition sub-module identifies the heart rate change amplitude and time interval of the pet within a continuous time segment, and uses the formula:

[0066]

[0067] Calculate the heart rate change rate value, and obtain the heart rate fluctuation rate trend according to the time distribution;

[0068] Among them, R v represents the heart rate change rate value, and H iThe heart rate value representing the i-th time point, H i+1 The heart rate value representing the (i + 1)-th time point, T i The time value corresponding to the i-th time point, T i+1 The time value corresponding to the (i + 1)-th time point, where n is the number of time points;

[0069] Explanation of formula parameters:

[0070] H i Represents the heart rate value at the i-th time point, with the unit of bpm;

[0071] T i Represents the time at the i-th time point, with the unit of minute;

[0072] |H i+1 -H i | Represents the absolute value of the heart rate difference between adjacent time points;

[0073] Calculation process of numerical example:

[0074] Extract the heart rate change amplitude and time interval within a continuous time segment. Taking the heart rate data as 80, 85, 83, 90, 95 bpm and the time interval as 1 minute as an example, the heart rate change amplitude is calculated as the absolute value of the heart rate difference between adjacent time points, that is, |85 - 80| = 5 bpm, |83 - 85| = 2 bpm, |90 - 83| = 7 bpm, |95 - 90| = 5 bpm. At the same time, the corresponding time intervals are all 1 minute, that is, the heart rate sequence is 80, 85, 83, 90, 95 bpm, and the corresponding time is 1, 2, 3, 4, 5 minutes. The calculation process is as follows:

[0075] Total sum of heart rate change amplitude:

[0076] |85 - 80| + |83 - 85| + |90 - 83| + |95 - 90| = 5 + 2 + 7 + 5 = 19;

[0077] Sum of squares of time intervals:

[0078] 1 2 +1 2 +1 2 +1 2 = 4;

[0079] Square of the total sum of heart rate differences:

[0080] (5 + 2 + 7 + 5) 2 = 19 2 = 361;

[0081] Square root operation of the denominator:

[0082]

[0083] Calculation of heart rate change rate value:

[0084]

[0085] The results show that the heart rate change rate value is 0.9945. Combining this rate value with the time distribution, a heart rate fluctuation rate trend graph can be drawn to obtain the heart rate fluctuation rate trend. The results indicate that the pet's heart rate fluctuates frequently during the monitored time period, but the fluctuation amplitude is small, indicating that the heart rate change is relatively stable.

[0086] The heart rate fluctuation stability evaluation sub-module uses the heart rate fluctuation rate trend to identify continuous heart rate fluctuation segments, identify the rate differences between adjacent segments, judge abnormal fluctuation points, and generate heart rate fluctuation characteristics;

[0087] Identify continuous heart rate fluctuation segments. Set the fluctuation trend graph to show significant fluctuations at 5 minutes, 15 minutes, and 25 minutes. Judge the stability of heart rate fluctuations by calculating the rate differences between adjacent segments. The calculation method of the rate difference is the absolute value of the difference between the heart rate change rate values in two adjacent time periods. Set the rate values at 5 minutes and 15 minutes to be 0.9945 and 0.9500 respectively, then the difference is |0.9945 - 0.9500| = 0.0445. If the set stability threshold is 0.05, when the difference value is less than 0.05, it is determined to be a stable fluctuation, otherwise it is unstable. According to the stability judgment of different time periods, abnormal fluctuation points are identified. Set the rate value at 25 minutes to be 0.9000, and the difference is |0.9500 - 0.9000| = 0.0500, which is equal to the set threshold, then it is determined to be in a boundary state, and the number and location of abnormal fluctuation points need to be further analyzed. By calculating the ratio of the stable fluctuation interval to the unstable interval, the heart rate fluctuation stability is calculated as 75% for the stable time ratio and 25% for the unstable time ratio, and the heart rate fluctuation characteristics are obtained.

[0088] Please refer to Figure 3 , and the specific steps for obtaining the motion mode trajectory are as follows:

[0089] The step frequency recognition sub-module synchronously collects the acceleration data during the pet's movement based on the heart rate fluctuation characteristics, identifies the changes in acceleration in different time intervals, and uses the formula:

[0090]

[0091] Calculate the step frequency and direction change value, and make a partition judgment based on the step frequency and direction change value to obtain the step frequency and motion direction change situation;

[0092] Among them, F bf represents the step frequency and direction change value, and Represent the accelerations in the x-axis and y-axis directions at different time points respectively. A is the number of acceleration measurements within the time interval, and v b represents the velocity within the different time intervals, and θ b is the angle of the motion direction corresponding to the velocity. B is the number of velocity measurements, and H c is the heart rate at the different time points, is the average heart rate, and C is the number of heart rate measurements;

[0093] Description of the parameter acquisition and assignment process:

[0094] Acceleration data: During the 30-minute exercise, it is sampled 10 times per second, A = 18000 times, and the acceleration values in the x-axis and y-axis directions range from 0.5 m / s² to 1.5 m / s²;

[0095] Velocity data: Obtained through GPS measurement, the velocity ranges from 1.0 m / s to 1.8 m / s at different time periods, and B = 1800 times;

[0096] Direction angle: Through the vector analysis of the position information, the direction angle ranges from 0° to 90°;

[0097] Heart rate data: Sampled once every 5 seconds, C = 360 times, the heart rate ranges from 110 bpm to 150 bpm, and the average heart rate

[0098] Calculation process of the formula example:

[0099] Calculation of the acceleration difference part:

[0100]

[0101] Calculation of the product of velocity and direction:

[0102]

[0103] Calculation of the heart rate fluctuation part:

[0104]

[0105] Integrated calculation:

[0106]

[0107] F bf ≈36 + 2 = 38;

[0108] The results show that during the movement, when the pet has a high heart rate fluctuation range and significant acceleration changes, the step frequency and direction change value is 38. Combining with the preset step frequency and direction change reference value range (30 to 45), it indicates that the pet's movement gait is in a high-change state during this time interval, meeting the determination criteria for the significant interval of step frequency and movement direction changes, and can provide accurate input data for subsequent gait stability analysis.

[0109] The gait stability analysis sub-module identifies the step frequency fluctuation range and the movement direction change trend through the step frequency and movement direction changes, identifies the gait time stability interval under different movement states, and generates a movement pattern trajectory.

[0110] According to the obtained step frequency and direction change values, filter out the interval where the step frequency change rate is between 0.2 Hz and 0.5 Hz. It is set that during the movement, the step frequency change rate of the pet is 0.3 Hz within the 15th to 20th minutes, which meets the screening conditions. Statistically analyze the direction change data within this time interval. The direction change calculates the vector angle between adjacent time points through the position information per second. It is set that within the 15th to 20th minutes, the average direction change angle is 30 degrees, then this interval is marked as a stable movement direction interval. Then, combined with the heart rate fluctuation characteristics, analyze the relationship between the heart rate and step frequency changes. It is set that the heart rate fluctuation is 6 bpm within this time interval, which is in the medium fluctuation range. Use time series analysis to perform sliding window calculations on the step frequency, direction change, and heart rate fluctuation data. Each window size is 1 minute, and the sliding interval is 10 seconds. Within each sliding window, calculate the step frequency standard deviation, direction change variance, and heart rate fluctuation mean. It is set that within a sliding window, the step frequency standard deviation is 0.05 Hz, the direction change variance is 0.1, and the heart rate fluctuation mean is 5 bpm. Then, the gait within this window is determined to be in a stable state. Through the statistical analysis of all sliding windows, filter out the continuous stable state intervals and generate a movement pattern trajectory.

[0111] Please refer to Figure 4 , and the specific steps for obtaining the abnormal state recognition result are as follows:

[0112] The movement pattern comparison sub-module uses the movement pattern trajectory to extract the pet's heart rate data under different movement states, compares the heart rate fluctuations under different movement states, calculates the difference between the heart rate fluctuation value in the movement state and the heart rate reference value in the static state, and generates a heart rate fluctuation difference value.

[0113] Through the movement tracking device worn by the pet, record the pet's position information, speed, and acceleration at different time periods. It is set that during the period from 8 am to 9 am, record the pet's walking trajectory at home, with a speed of 0.5 m / s and an acceleration of 0.2 m / s 2, during the period from 9 o'clock to 10 o'clock, record the trajectory of the pet running outdoors, with a speed of 1.5 meters per second and an acceleration of 0.8 meters per second 2 , for the heart rate fluctuation situation in the motion state, call the heart rate data collected in real time of the pet in different states, set the heart rate at rest to 80 beats per minute, the walking state to 100 beats per minute, and the running state to 140 beats per minute. Calculate the differences between the heart rate fluctuation values in each motion state and the heart rate benchmark value at rest, and the differences are 20 beats per minute and 60 beats per minute respectively. According to the differences for comparison, identify whether the heart rate fluctuation is within the normal range. Set the normal range of heart rate fluctuation at rest to ±10 beats per minute. Through comparison, it is found that the heart rate fluctuation in the walking state exceeds the normal range by 10 beats per minute, and the heart rate fluctuation in the running state exceeds the normal range by 50 beats per minute, and obtain the heart rate fluctuation difference value.

[0114] Based on the heart rate fluctuation difference value, the heart rate change interval analysis sub-module combines the heart rate data in the motion state and the heart rate data at rest to analyze the change situation of the heart rate interval in the motion state and the heart rate interval at rest, and uses the formula:

[0115]

[0116] Calculate the difference value of the heart rate change interval to obtain the heart rate fluctuation interval range;

[0117] Among them, H d represents the difference value of the heart rate change interval, H mf represents the heart rate value in the f-th motion state, H sf represents the heart rate value in the f-th rest state, represents the mean value of the heart rate values in the motion state, min(H m ) represents the minimum value of the heart rate in the motion state, max(H s ) represents the maximum value of the heart rate in the rest state, and F represents the total number of sample data;

[0118] Actual operation data acquisition and interpretation:

[0119] Set in the test scenario, 5 groups of heart rate data of the pet at rest are collected as 80, 82, 81, 79, 80 beats per minute;

[0120] In the motion state, they are 100, 102, 98, 101, 99 beats per minute;

[0121] All data are obtained in real time through a motion tracking device combined with a heart rate monitoring device, and the total number of samples F = 5;

[0122] Formula operation process:

[0123] Calculate the sum of absolute differences:

[0124] |100 - 80| + |102 - 82| + |98 - 81| + |101 - 79| + |99 - 80| = 98;

[0125] Mean of absolute differences:

[0126]

[0127] Variance part:

[0128] (100 - 100) 2 + (102 - 100) 2 + (98 - 100) 2 + (101 - 100) 2 + (99 - 100) 2 = 10;

[0129]

[0130] Mean of heart rate intervals:

[0131]

[0132] Comprehensive calculation:

[0133] H d = 19.6 + 1.41 - 90 = -69.0;

[0134] This result indicates that the heart rate fluctuation of the pet during exercise exceeds the set normal range. The normal fluctuation range is -20 to +20 beats per minute, while the actual value is -69.0 beats per minute, indicating an abnormal state that needs to be further confirmed and an abnormal state identification is required.

[0135] The fluctuation situation identification sub-module uses the heart rate fluctuation interval range to compare the real-time heart rate fluctuation with the set normal threshold of heart rate fluctuation, judge the heart rate fluctuation situation exceeding the threshold, and generate an abnormal state identification result;

[0136] Based on the situation that the heart rate fluctuation exceeds the normal range, it is necessary to set a normal threshold for heart rate fluctuation. This threshold is comprehensively set according to the pet's body size, age, and previous health status data. Through the analysis of data of pets of the same breed, the normal threshold for heart rate fluctuation is set to -20 to +20 beats per minute. Among them, -20 beats per minute represents the lower limit of heart rate change below the resting state, and +20 beats per minute represents the upper limit of heart rate change above the resting state. It is set that when the pet is in a mild exercise state, the heart rate fluctuation remains at 10 to 15 beats per minute; in a moderate exercise state, the heart rate fluctuation is between 20 and 30 beats per minute; and in a strenuous exercise state, the heart rate fluctuation will reach 40 to 60 beats per minute. Therefore, the normal threshold needs to be set according to the average level of the pet individual, and the setting range is -20 to +20 beats per minute. Compare the heart rate fluctuation interval difference of -69.0 beats per minute with the normal threshold, and calculate the difference beyond the range, that is, -69.0 - (-20) = -49.0 beats per minute, indicating that it exceeds the normal lower limit by 49 beats per minute. To further confirm whether this fluctuation is an abnormal state, it is necessary to analyze the duration of heart rate fluctuation. Set the duration of this abnormal fluctuation to 15 minutes, while under normal circumstances, the duration of heart rate fluctuation of the pet in a strenuous exercise state does not exceed 5 minutes. This duration also significantly exceeds the normal range. It is also necessary to consider whether this fluctuation is related to changes in the external environment. In a high-temperature environment, the heart rate fluctuation of the pet will naturally increase. Set the ambient temperature at that time to 25 degrees Celsius, which is within the normal range. Therefore, exclude the interference of environmental factors. Comprehensively analyze the heart rate fluctuation difference, duration, and environmental factors to judge that the current heart rate fluctuation of the pet is in an abnormal state and generate an abnormal state recognition result.

[0137] Please refer to Figure 5 , and the specific steps for obtaining the heart rate abnormality grading result are as follows:

[0138] The exercise state analysis sub-module uses the abnormal state recognition result to record the exercise time and interval time before and after the occurrence of the abnormal heart rate of the pet, extracts the change trend of the exercise state, judges whether the abnormal heart rate is caused by exercise, and calculates the exercise duration ratio corresponding to the abnormal heart rate. The formula is used:

[0139]

[0140] Calculate the exercise state abnormal correlation rate;

[0141] Among them, R m represents the exercise state abnormal correlation rate, T mg represents the g-th exercise duration before the occurrence of the abnormal heart rate, A mg represents the g-th exercise intensity before the occurrence of the abnormal heart rate, T bg represents the g-th exercise interval time after the occurrence of the abnormal heart rate, and G represents the total amount of exercise data corresponding to the abnormal heart rate;

[0142] Detailed Explanation of the Formula and Calculation Derivation Process:

[0143] T mg : Represents the exercise duration before the g-th abnormal heart rate occurs (unit: minutes). This data can be obtained by wearing a motion sensor such as an accelerometer on the pet in real-time. Suppose the exercise durations of a certain pet within a day are 12 minutes, 15 minutes, and 10 minutes respectively;

[0144] A mg : Represents the exercise intensity before the g-th abnormal heart rate occurs. The exercise intensity is measured by metabolic equivalents (METs). 1 MET is approximately equal to consuming 3.5 milliliters of oxygen per kilogram of body weight per minute. The exercise intensity can be determined by methods such as cardiopulmonary exercise testing (CPET) according to the type and speed of the pet's activities. Suppose the intensities of the above three exercises are 4 METs, 5 METs, and 3.5 METs respectively;

[0145] T bg : Represents the interval time after the g-th abnormal heart rate occurs (unit: minutes). This can be determined by recording the duration of the resting or low-activity state after the abnormal heart rate event using a heart rate monitoring device. The interval times after the three abnormal heart rate events are 5 minutes, 7 minutes, and 6 minutes respectively;

[0146] G: Represents the total number of abnormal heart rate events during the monitoring period. In this example, G = 3;

[0147] Calculation Process:

[0148] Calculate the numerator part:

[0149]

[0150] Calculate the denominator part:

[0151]

[0152] Calculate R m :

[0153]

[0154] The results show that there is a significant association between the abnormal heart rate of the pet and the exercise state. The m higher the R value, the longer the duration and higher the intensity of the exercise the pet experiences before the abnormal heart rate occurs, indicating that exercise is one of the main factors triggering the abnormal heart rate. In practical applications, this index can be used to evaluate the impact of the pet's exercise on heart health, formulate a reasonable exercise plan, and prevent potential heart problems.

[0155] The heart rate abnormality grading sub-module identifies the heart rate fluctuation trends corresponding to different abnormal types based on the abnormal correlation rate of the exercise state, counts the abnormal fluctuation frequency and duration, divides the heart rate abnormality levels, and obtains the heart rate abnormality grading result;

[0156] Combined with the abnormal state recognition result, analyze the heart rate fluctuation trends corresponding to different abnormal types. For different types such as tachycardia, bradycardia, and arrhythmia, respectively count their occurrence frequencies and durations. Set that in a one-week monitoring, tachycardia occurs 5 times, each lasting 2 minutes; bradycardia occurs 3 times, each lasting 1.5 minutes; arrhythmia occurs 2 times, each lasting 3 minutes. Count the abnormal fluctuation frequency and duration, summarize the above data, calculate the total duration of each abnormal type. The total duration of tachycardia is 10 minutes, bradycardia is 4.5 minutes, and arrhythmia is 6 minutes. Divide the heart rate abnormality levels. According to the total duration and occurrence frequency, set thresholds to divide the abnormal degree into mild, moderate, and severe. Those with a total duration less than 5 minutes and an occurrence frequency less than 3 times are mild, 5 - 10 minutes and 3 - 5 times are moderate, and greater than 10 minutes or more than 5 times are severe. Combine the abnormal correlation rate of the exercise state, comprehensively evaluate the heart rate abnormality situation of the pet, and obtain the heart rate abnormality grading result.

[0157] Please refer to Figure 6 , and the steps for obtaining the abnormal warning result of heart rate monitoring are specifically as follows:

[0158] The abnormal signal recognition sub-module uses the heart rate abnormality grading result to obtain the heart rate data corresponding to the abnormal state, combines the heart rate deviation amplitude, fluctuation frequency, and duration, and uses the formula:

[0159]

[0160] Calculate the abnormal signal strength, compare it with the trigger threshold, screen out the abnormal states that meet the over-limit standard, and generate the abnormal signal strength screening result;

[0161] Among them, AS represents the abnormal signal strength, H represents the real-time heart rate value, H ref represents the reference heart rate value, D o represents the heart rate fluctuation deviation value, Q represents the number of heart rate samples within the calculation window, W represents the weight parameter, T var represents the heart rate time variability;

[0162] It is necessary to determine the source of abnormal heart rate data. In practical applications, heart rate data can be real-time monitored through wearable devices, including smart pet collars or smart biological monitoring devices. The devices can record heart rate data at a certain sampling frequency (set to once per second) and transmit it to the data processing system via wireless communication. Among the collected heart rate data, it is necessary to screen out the heart rate records that meet the abnormal grading criteria. This process can be achieved by setting a reference heart rate range. For canine pets, the resting heart rate in a healthy state is between 60 and 140 bpm, while for feline pets, the resting heart rate is between 140 and 220 bpm. If the heart rate record at a certain moment exceeds the normal heart rate range, it is marked as abnormal data. Calculate based on the heart rate deviation amplitude, fluctuation frequency, and duration to determine the abnormal signal strength;

[0163] Parameter meaning:

[0164] H represents the current heart rate value (unit: bpm). Set the current heart rate data to 180 bpm;

[0165] H ref represents the reference heart rate value (unit: bpm). Set the reference heart rate value for canine pets to 120 bpm;

[0166] Q represents the number of heart rate samples within the calculation window. Set Q to 6;

[0167] W represents the fluctuation weight parameter, set according to the influence of heart rate fluctuations. Set it to 1.2;

[0168] T var represents the heart rate time variability. Calculate the variance of the heart rate changes within this time window. Set the calculated variance value to 25;

[0169] D o represents the heart rate fluctuation deviation value (unit: bpm). The heart rate deviation amplitude can be calculated by the absolute difference between the current heart rate value and the reference heart rate value. Set if the heart rate data of a certain canine pet is 180 bpm and its reference heart rate value is 120 bpm, then the deviation amplitude is calculated as follows:

[0170] |180 - 120| = 60 bpm;

[0171] To calculate the heart rate fluctuation deviation value, a certain time window needs to be selected (set to 1 minute). Within this time window, count the changes in heart rate data points. If the heart rate data within this time window is 160, 170, 175, 180, 190, 185, 195 bpm, then calculate its fluctuation deviation mean value:

[0172]

[0173] Substitute the above values into the formula for calculation:

[0174]

[0175] AS = 60 + 7.5×1.2 + 5 = 60 + 9 + 5 = 74;

[0176] The results show that the abnormal signal strength does not reach the trigger threshold (the set threshold is 80 bpm), so this situation will not trigger a remote warning.

[0177] The remote warning push sub-module uses the screening result of the abnormal signal strength to judge whether the remote warning condition is met. If it is met, a remote abnormal warning signal is pushed, the abnormal status is recorded, and a heart rate monitoring abnormal warning result is generated.

[0178] Judge whether the remote warning condition is met. If it is met, a remote abnormal warning signal is pushed, and the abnormal status is recorded. During the remote warning push process, a trigger threshold needs to be set, that is, when the abnormal signal strength AS exceeds a certain critical value, a warning is triggered. Set this trigger threshold to 80 bpm. If the abnormal signal strength is greater than 80 bpm, the system will automatically generate an abnormal alarm and send it to the remote server through wireless communication technology. At the same time, record and store this abnormal status for subsequent analysis. Set that when the abnormal signal is triggered, the heart rate status of the pet is 195 bpm and the abnormal signal strength is 82 bpm, then the alarm condition is met, the system pushes the alarm, and a heart rate monitoring abnormal warning result is generated.

[0179] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A pet heart rate monitoring system based on cloud computing, characterized in that, The system includes: The heart rate fluctuation recording module collects the heart rate data of mammalian pets, identifies the amplitude of heart rate changes within a short period of time, calculates the heart rate change rate, analyzes multiple consecutive heart rate fluctuation segments, and generates heart rate fluctuation characteristics; The motion pattern tracking module uses the heart rate fluctuation characteristics to collect the acceleration data of the pet, analyzes the change of step frequency and motion direction, identifies the stability of the gait during exercise, and generates a motion pattern trajectory; The abnormal marking module, through the motion pattern trajectory, compares the heart rate fluctuations of the pet under different motion states, identifies the corresponding relationship between the heart rate and the motion pattern, analyzes the heart rate change intervals during motion and at rest, records the situation where the heart rate fluctuation exceeds the normal range, marks the abnormal state, and generates an abnormal state recognition result; The grading evaluation module uses the abnormal state recognition result to analyze the duration and interruption of the pet's exercise when the heart rate abnormality occurs, divides different grades, and obtains the heart rate abnormality grading result; The abnormal signal triggering module, according to the heart rate abnormality grading result, judges whether the pet's heart rate abnormality exceeds the triggering threshold. If it exceeds, it pushes a remote abnormal warning signal to obtain the heart rate monitoring abnormal warning result.

2. The pet heart rate monitoring system based on cloud computing according to claim 1, characterized in that, The heart rate fluctuation characteristics include the peak value, valley value, average heart rate, and heart rate fluctuation frequency of the heart rate. The motion pattern trajectory includes step frequency statistics, motion trajectory mapping, and gait stability analysis results. The abnormal state recognition result includes the timestamp of heart rate abnormality, abnormal duration, and abnormal heart rate interval. The heart rate abnormality grading result is the analysis result of the heart rate fluctuation trend, abnormal severity rating, and potential factors affecting motion. The heart rate monitoring abnormal warning result includes the alarm triggering threshold setting, alarm sending time, and alarm status update information.

3. The pet heart rate monitoring system based on cloud computing according to claim 1, characterized in that The specific steps for obtaining the heart rate fluctuation characteristics are as follows: The heart rate signal acquisition sub-module collects the heart rate data of the pet, detects the continuous heart rate of the pet, screens complete heart rate segments, deletes abnormal data, and generates a heart rate data sequence; The heart rate change amplitude recognition sub-module, based on the heart rate data sequence, identifies the heart rate change amplitude and time interval of the pet within a continuous time segment, and uses the formula: Calculate the heart rate change rate value, and obtain the heart rate fluctuation rate trend according to the time distribution; Wherein, R v represents the heart rate change rate value, H i represents the heart rate value at the i-th time point, H i+1 represents the heart rate value at the (i + 1)-th time point, T i represents the time value corresponding to the i-th time point, T i+1 represents the time value corresponding to the (i + 1)-th time point, and n is the number of time points; The heart rate fluctuation stability evaluation sub-module uses the heart rate fluctuation rate trend to identify continuous heart rate fluctuation segments, identifies the rate difference between adjacent segments, judges abnormal fluctuation points, and generates heart rate fluctuation characteristics.

4. The pet heart rate monitoring system based on cloud computing according to claim 3, characterized in that, The specific steps for obtaining the motion pattern trajectory are as follows: The step frequency recognition sub-module, based on the heart rate fluctuation characteristics, synchronously collects the acceleration data during the pet's exercise, identifies the changes in acceleration within different time intervals, and uses the formula: Calculate the step frequency and direction change value, and make a partition judgment according to the step frequency and direction change value to obtain the step frequency and motion direction change situation; Among them, F bf represents the step frequency and direction change value, and respectively represent the accelerations in the x-axis and y-axis directions at the differential time points, A is the number of acceleration measurements within the time interval, v b represents the speed within the differential time interval, θ b is the motion direction angle corresponding to the speed, B is the number of speed measurements, H c is the heart rate at the differential time point, is the average heart rate, and C is the number of heart rate measurements; The gait stability analysis sub-module, through the step frequency and motion direction change situation, identifies the step frequency fluctuation interval and the motion direction change trend, and identifies the gait time stability interval under different motion states to generate a motion pattern trajectory.

5. The pet heart rate monitoring system based on cloud computing according to claim 4, wherein The specific steps for obtaining the abnormal state recognition result are as follows: The motion mode comparison sub-module uses the motion mode trajectory to extract the pet heart rate data under different motion states, compares the heart rate fluctuations under different motion states, calculates the difference between the heart rate fluctuation value under the motion state and the heart rate reference value under the static state, and generates a heart rate fluctuation difference value; The heart rate change interval analysis sub-module analyzes the changes in the heart rate interval under the motion state and the heart rate interval under the static state based on the heart rate fluctuation difference value, combined with the heart rate data under the motion state and the heart rate data under the static state, using the formula: Calculate the difference in the heart rate change interval to obtain the heart rate fluctuation interval range; Among them, H d represents the difference in the heart rate variation range, H mf represents the heart rate value in the f-th exercise state, H sf represents the heart rate value in the f-th resting state, represents the mean value of the heart rate values in the exercise state, min(H m ) represents the minimum heart rate in the exercise state, max(H s ) represents the maximum heart rate in the resting state, and F represents the total number of sample data; The fluctuation situation identification sub-module uses the heart rate fluctuation interval range to compare the real-time heart rate fluctuation with the set normal heart rate fluctuation threshold, judges the heart rate fluctuation situation exceeding the threshold, and generates an abnormal state identification result.

6. The pet heart rate monitoring system based on cloud computing according to claim 5, characterized in that, The specific steps for obtaining the heart rate abnormality grading result are as follows: The motion state analysis sub-module uses the abnormal state identification result to record the motion time and the interval time before and after the occurrence of the abnormal heart rate of the pet, extracts the change trend of the motion state, judges whether the abnormal heart rate is caused by exercise, and calculates the motion duration ratio corresponding to the abnormal heart rate, using the formula: Calculate the abnormal association rate of the motion state; Among them, R m represents the abnormal motion state correlation rate, T mg represents the g-th exercise duration before the occurrence of abnormal heart rate, A mg represents the g-th exercise intensity before the occurrence of abnormal heart rate, T bg represents the g-th exercise interval time after the occurrence of abnormal heart rate, and G represents the total amount of exercise data corresponding to the abnormal heart rate; The heart rate abnormality grading sub-module based on the abnormal association rate of the motion state identifies the heart rate fluctuation trend corresponding to different abnormal types, counts the abnormal fluctuation frequency and duration, divides the heart rate abnormality level, and obtains the heart rate abnormality grading result.

7. The pet heart rate monitoring system based on cloud computing according to claim 6, wherein The specific steps for obtaining the heart rate monitoring abnormal warning result are as follows: The abnormal signal identification sub-module uses the heart rate abnormality grading result to obtain the heart rate data corresponding to the abnormal state, combined with the heart rate deviation amplitude, fluctuation frequency and duration, using the formula: Calculate the abnormal signal intensity, compare it with the trigger threshold, screen the abnormal states that meet the over-limit standard, and generate an abnormal signal intensity screening result; Among them, AS represents the abnormal signal intensity, H represents the real-time heart rate value, and H ref represents the reference heart rate value, D o represents the heart rate fluctuation deviation value, Q represents the number of heart rate samples within the calculation window, W represents the weight parameter, and T var represents the heart rate temporal variability; The remote warning push sub-module uses the abnormal signal intensity screening result to judge whether the remote warning condition is met. If it is met, it pushes a remote abnormal warning signal and records the abnormal state, generating a heart rate monitoring abnormal warning result.

Images