Physical data acquisition method and system based on BLE technology

Abstract

The invention discloses a physical ability data acquisition method and system based on a BLE technology, relates to the technical field of intelligent data acquisition, and can acquire heart rate data of a monitored student object in real time through an intelligent wearable device, and transmit the heart rate data to a cloud platform by using the BLE technology. The heart rate data of the monitored student object can be subjected to time sequence feature analysis and multi-scale association by using a data processing and analysis algorithm based on artificial intelligence in a cloud platform; therefore, whether a heart rate abnormity early warning prompt is generated or not is intelligently judged according to the multi-scale heart rate mode characteristics of the heart rate data of the monitored student object. Through the mode, the real-time performance and accuracy of data can be ensured by utilizing the characteristics of low power consumption and efficient transmission of the BLE technology, and meanwhile, the system can quickly and accurately identify the abnormal heart rate mode and send out an early warning prompt when necessary, so that false alarms and missing alarms are reduced, and timely intervention and support are provided for health management of students.

Description

Technical Field

[0001] The present application relates to the field of intelligent data collection technology, and more specifically, to a physical fitness data collection method and system based on BLE technology. Background Art

[0002] Physical data collection methods refer to the process of using various technologies and equipment to collect and record individual physiological and athletic performance data during sports activities or daily activities. These data usually include heart rate, steps, speed, distance, calories burned, sleep quality and other indicators, which can be used to assess individual health status, exercise capacity, recovery status and potential health risks.

[0003] Heart rate is one of the important indicators that reflect heart function and overall health. By monitoring heart rate, health problems such as heart disease and arrhythmia can be discovered in time, helping to prevent and treat cardiovascular diseases. During exercise and fitness, monitoring heart rate can help individuals understand their exercise intensity and endurance level, avoid over- or under-exercise, and improve exercise effectiveness and safety.

[0004] However, traditional heart rate data collection and monitoring methods are usually based on fixed thresholds or rules for judgment, lacking personalization and dynamism, and cannot fully consider individual physiological characteristics and changes. In addition, methods based on fixed thresholds are easily affected by individual physiological differences, environmental factors, etc., resulting in a high misdiagnosis rate, and normal variations may be misjudged as abnormal. In addition, traditional methods usually require manual intervention or regular measurements, and cannot achieve continuous and real-time heart rate monitoring, which limits the timely detection and treatment of heart rate abnormalities.

[0005] Therefore, an optimized method for collecting physical performance data is desired. Summary of the invention

[0006] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a physical data collection method and system based on BLE technology, which can collect the heart rate data of the monitored student object in real time through a smart wearable device, and transmit the heart rate data to a cloud platform using BLE technology, so as to use artificial intelligence-based data processing and analysis algorithms in the cloud platform to perform time series feature analysis and multi-scale association on the heart rate data of the monitored student object, so as to intelligently determine whether to generate an abnormal heart rate warning prompt based on the multi-scale heart rate pattern characteristics of the heart rate data of the monitored student object. In this way, the low power consumption and efficient transmission characteristics of BLE technology can be used to ensure the real-time and accuracy of the data. At the same time, the system can quickly and accurately identify abnormal heart rate patterns and issue warning prompts when necessary, reducing false alarms and missed alarms, thereby providing timely intervention and support for students' health management.

[0007] According to one aspect of the present application, a method for collecting physical fitness data based on BLE technology is provided, which includes:

[0008] Collect the time series of heart rate data of monitored student subjects through smart wearable devices;

[0009] Upload the time series of the heart rate data to the cloud platform through the BEL module;

[0010] On the cloud platform, arranging the time series of the heart rate data into a heart rate time series input vector according to the time dimension;

[0011] On the cloud platform, performing vector segmentation on the heart rate timing input vector to obtain a sequence of heart rate local timing input vectors, and then performing heart rate timing pattern feature extraction on the sequence of heart rate local timing input vectors to obtain a sequence of heart rate local timing associated feature vectors;

[0012] Extracting local time domain correlation features of the heart rate from the sequence of local time series correlation feature vectors of the heart rate to obtain a first time scale heart rate pattern feature vector;

[0013] Extracting the global time domain correlation features of the heart rate from the sequence of the local time series correlation feature vectors of the heart rate to obtain a second time scale heart rate pattern feature vector;

[0014] Using a dynamic element-level feature fusion module to fuse the first time-scale heart rate pattern feature vector and the second time-scale heart rate pattern feature vector to obtain a multi-scale heart rate pattern feature vector as a multi-scale heart rate pattern feature;

[0015] Based on the multi-scale heart rate pattern characteristics, an early warning result is obtained.

[0016] According to another aspect of the present application, a physical fitness data collection system based on BLE technology is provided, which includes:

[0017] A heart rate data acquisition module is used to collect the time series of heart rate data of the monitored student object through a smart wearable device;

[0018] A heart rate data uploading module, used to upload the time series of the heart rate data to the cloud platform through the BEL module;

[0019] A heart rate data time series arrangement module, used for arranging the time series of the heart rate data into a heart rate time series input vector according to the time dimension on the cloud platform;

[0020] A vector segmentation feature extraction module is used to perform vector segmentation on the heart rate timing input vector on the cloud platform to obtain a sequence of heart rate local timing input vectors, and then perform heart rate timing pattern feature extraction on the sequence of heart rate local timing input vectors to obtain a sequence of heart rate local timing associated feature vectors;

[0021] A local time domain correlation module, used for extracting the local time domain correlation features of the heart rate from the sequence of the local time series correlation feature vectors of the heart rate to obtain a first time scale heart rate pattern feature vector;

[0022] A global time domain association module, used for extracting the global time domain association features of the heart rate from the sequence of the local time series association feature vectors of the heart rate to obtain a second time scale heart rate pattern feature vector;

[0023] a feature fusion module, configured to use a dynamic element-level feature fusion module to fuse the first time-scale heart rate pattern feature vector and the second time-scale heart rate pattern feature vector to obtain a multi-scale heart rate pattern feature vector as a multi-scale heart rate pattern feature;

[0024] The early warning result generating module is used to obtain the early warning result based on the multi-scale heart rate pattern characteristics.

[0025] The present application has at least the following technical effects: Compared with the prior art, the present application provides a method and system for collecting physical data based on BLE technology, which can collect the heart rate data of the monitored student object in real time through smart wearable devices, and transmit the heart rate data to the cloud platform using BLE technology, so as to use artificial intelligence-based data processing and analysis algorithms in the cloud platform to perform time series feature analysis and multi-scale association on the heart rate data of the monitored student object, so as to intelligently determine whether to generate an abnormal heart rate warning prompt based on the multi-scale heart rate pattern characteristics of the heart rate data of the monitored student object. In this way, the low power consumption and efficient transmission characteristics of BLE technology can be used to ensure the real-time and accuracy of the data. At the same time, the system can quickly and accurately identify abnormal heart rate patterns and issue warning prompts when necessary, reducing false alarms and missed alarms, thereby providing timely intervention and support for students' health management. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0027] Figure 1This is a flow chart of a method for collecting physical fitness data based on BLE technology according to an embodiment of the present application.

[0028] Figure 2 Schematic diagram of the architecture of a method for collecting physical fitness data based on BLE technology according to an embodiment of the present application.

[0029] Figure 3 This is a flowchart of the physical fitness data collection method based on BLE technology according to an embodiment of the present application, which performs vector segmentation on the heart rate timing input vector to obtain a sequence of heart rate local timing input vectors, and then performs heart rate timing pattern feature extraction on the sequence of heart rate local timing input vectors to obtain a sequence of heart rate local timing associated feature vectors.

[0030] Figure 4 It is a block diagram of a physical fitness data collection system based on BLE technology according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] The embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, but rather these embodiments are provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0032] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0033] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0034] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0035] Physical data collection methods refer to the process of using various technologies and equipment to collect and record individual physiological and athletic performance data during sports activities or daily activities. These data usually include heart rate, steps, speed, distance, calories burned, sleep quality and other indicators, which can be used to assess individual health status, exercise capacity, recovery status and potential health risks.

[0036] Heart rate is one of the important indicators that reflect heart function and overall health. By monitoring heart rate, health problems such as heart disease and arrhythmia can be discovered in time, helping to prevent and treat cardiovascular diseases. During exercise and fitness, monitoring heart rate can help individuals understand their exercise intensity and endurance level, avoid over- or under-exercise, and improve exercise effectiveness and safety.

[0037] However, traditional heart rate data collection and monitoring methods are usually based on fixed thresholds or rules for judgment, lacking personalization and dynamism, and cannot fully consider individual physiological characteristics and changes. In addition, methods based on fixed thresholds are easily affected by individual physiological differences, environmental factors, etc., resulting in a high misdiagnosis rate, and normal variations may be misjudged as abnormal. In addition, traditional methods usually require manual intervention or regular measurements, and cannot achieve continuous and real-time heart rate monitoring, which limits the timely detection and treatment of heart rate abnormalities.

[0038] Specifically, the BLE technology, namely Bluetooth Low Energy (low-power Bluetooth technology), is a subset of the Bluetooth technical specification and has been introduced since Bluetooth version 4.0. Specifically, compared with classic Bluetooth, the main feature of BLE is that it can provide sufficient data transmission rate while maintaining low power consumption, making it very suitable for devices in the field of Internet of Things (IoT), such as smart bracelets, smart light bulbs, door locks, etc. In particular, BLE technology is also widely used in physical data collection. Specifically, the application of BLE technology in physical data collection is mainly reflected in wearable devices and health monitoring devices. These devices can monitor and record users' physical data in real time, help users better understand their physical condition and exercise effects, and provide users with health management and exercise guidance, thereby improving the quality of life and health level.

[0039] Based on this, in the technical solution of the present application, a physical data collection method based on BLE technology is proposed, which can collect the heart rate data of the monitored student object in real time through a smart wearable device, and transmit the heart rate data to a cloud platform using BLE technology, so as to use artificial intelligence-based data processing and analysis algorithms in the cloud platform to perform time series feature analysis and multi-scale association on the heart rate data of the monitored student object, so as to intelligently determine whether to generate an abnormal heart rate warning prompt based on the multi-scale heart rate pattern characteristics of the heart rate data of the monitored student object. In this way, the low power consumption and efficient transmission characteristics of BLE technology can be used to ensure the real-time and accuracy of the data. At the same time, the system can quickly and accurately identify abnormal heart rate patterns and issue warning prompts when necessary, reducing false alarms and missed alarms, thereby providing timely intervention and support for students' health management.

[0040] Figure 1 This is a flow chart of a method for collecting physical fitness data based on BLE technology according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of the architecture of a method for collecting physical fitness data based on BLE technology according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to the physical fitness data collection method based on BLE technology of the embodiment of the present application, it includes: S110, collecting the time series of heart rate data of the monitored student object through the smart wearable device; S120, uploading the time series of the heart rate data to the cloud platform through the BEL module; S130, on the cloud platform, arranging the time series of the heart rate data into a heart rate timing input vector according to the time dimension; S140, on the cloud platform, performing vector segmentation on the heart rate timing input vector to obtain a sequence of heart rate local timing input vectors, and then performing heart rate timing pattern feature extraction on the sequence of heart rate local timing input vectors to obtain a heart rate local timing correlation feature vector. S150, extracting the local time-domain correlation features of the heart rate on the sequence of local time-series correlation feature vectors to obtain a first time-scale heart rate pattern feature vector; S160, extracting the global time-domain correlation features of the heart rate on the sequence of local time-series correlation feature vectors to obtain a second time-scale heart rate pattern feature vector; S170, using a dynamic element-level feature fusion module to fuse the first time-scale heart rate pattern feature vector and the second time-scale heart rate pattern feature vector to obtain a multi-scale heart rate pattern feature vector as a multi-scale heart rate pattern feature; and, S180, obtaining a warning result based on the multi-scale heart rate pattern feature.

[0041] In step S110, the time series of the heart rate data of the monitored student object is collected by the smart wearable device. It should be understood that, considering that the time series of the heart rate data contains a series of heartbeat events, such as the heart rate of the student at rest, tachycardia (heart rate is too fast) or bradycardia (heart rate is too slow), these situations can understand the health of their cardiovascular system or indicate potential health problems. Based on this, in the technical solution of the present application, the time series of the heart rate data of the monitored student object is collected by the smart wearable device, and the time series analysis and processing are performed on it, so that potential health problems, such as arrhythmia, heart disease, etc., can be discovered in time, and timely warnings can be given, so as to intervene and treat them as soon as possible.

[0042] In step S120, the time series of the heart rate data is uploaded to the cloud platform through the BEL module. Accordingly, the heart rate data can be transmitted from the smart wearable device to the cloud platform in real time through the BLE technology, so that users and medical professionals can monitor heart rate changes in real time and detect any abnormalities in time. In particular, BLE technology is specially designed for low-power applications, which means that even if data is continuously transmitted, it will not significantly affect the battery life of the smart wearable device, thereby ensuring the long-term use of the device. The cloud platform can integrate advanced data analysis tools and artificial intelligence algorithms to conduct in-depth analysis of heart rate data, thereby providing more accurate health assessments and personalized health recommendations.

[0043] In step S130, on the cloud platform, the time series of the heart rate data is arranged as a heart rate time series input vector according to the time dimension. It should be understood that, considering that the heart rate data has different heart rate changes at different times, such as heart rate changes at different times of the day, different days of the week, or before and after a specific activity, that is, the heart rate data has certain characteristic information on the time scale. Based on this, in order to better reveal the time series change trend and fluctuation of the heart rate of the monitored student object over a period of time, in the technical solution of the present application, on the cloud platform, the time series of the heart rate data is arranged as a heart rate time series input vector according to the time dimension, so that the heart rate data can be processed and analyzed, so as to understand the time series trend or abnormal state of the heart rate change in the heart rate time series input vector, thereby providing early warning for the health management of students so that medical measures can be taken in time.

[0044] In step S140, on the cloud platform, the heart rate timing input vector is vector segmented to obtain a sequence of heart rate local timing input vectors, and then the sequence of heart rate local timing input vectors is subjected to heart rate timing pattern feature extraction to obtain a sequence of heart rate local timing associated feature vectors. Figure 3The present invention is a flow chart of performing vector segmentation on the heart rate timing input vector to obtain a sequence of heart rate local timing input vectors, and then extracting heart rate timing pattern features from the sequence of heart rate local timing input vectors to obtain a sequence of heart rate local timing associated feature vectors in a physical fitness data collection method based on BLE technology according to an embodiment of the present application. Specifically, in an embodiment of the present application, Figure 3 As shown, on the cloud platform, after vector segmentation is performed on the heart rate timing input vector to obtain a sequence of heart rate local timing input vectors, heart rate timing pattern feature extraction is performed on the sequence of heart rate local timing input vectors to obtain a sequence of heart rate local timing associated feature vectors, including: S210, on the cloud platform, vector segmentation is performed on the heart rate timing input vector based on a predetermined time scale to obtain a sequence of heart rate local timing input vectors; and, S220, on the cloud platform, the sequence of heart rate local timing input vectors is passed through a heart rate timing pattern feature extractor based on a one-dimensional convolutional layer to obtain a sequence of heart rate local timing associated feature vectors.

[0045] Specifically, in step S210, on the cloud platform, the heart rate time series input vector is vector segmented based on a predetermined time scale to obtain a sequence of the heart rate local time series input vector. Accordingly, considering that the heart rate data usually contains a large number of time points, directly processing the entire heart rate time series may result in high computational complexity and long processing time, and the heart rate time series input vector has different change trends on different time scales, that is, the time series changes and fluctuations of the heart rate may only exist on a shorter local time scale, and are not obvious on the entire time scale. Therefore, in order to analyze and process the characteristic information about the heart rate in each local time scale in a more fine-grained manner, in the technical solution of the present application, on the cloud platform, the heart rate time series input vector is vector segmented based on a predetermined time scale to obtain a sequence of the heart rate local time series input vector, so that a whole heart rate time series input vector containing a large number of time periods can be decomposed into a smaller and more refined sequence of the heart rate local time series input vector, so that the heart rate characteristics in a specific time period can be captured, so as to better understand the pattern and trend of the heart rate change in the local time period, thereby improving the accuracy of the subsequent warning results.

[0046] Specifically, in step S220, on the cloud platform, the sequence of the heart rate local time series input vectors is passed through a heart rate time series pattern feature extractor based on a one-dimensional convolutional layer to obtain a sequence of the heart rate local time series associated feature vectors. It should be understood that, considering that the sequence of the heart rate local time series input vectors has characteristic information about the heart rate in the local time series, and the one-dimensional convolutional layer is used in deep learning to automatically learn and extract features from time series data, it can identify patterns and structures within the local time window, which is very useful for understanding the local time series changes in the heart rate time series data. Therefore, in the technical solution of the present application, on the cloud platform, the sequence of the heart rate local time series input vectors is passed through a heart rate time series pattern feature extractor based on a one-dimensional convolutional layer to respectively capture and mine out the local time series feature information about the heart rate in each heart rate local time series input vector, and obtain a more reliable sequence of heart rate local time series associated feature vectors.

[0047] In step S150, the sequence of the local temporal correlation feature vectors of the heart rate is subjected to the local temporal correlation feature extraction of the heart rate to obtain the first time scale heart rate pattern feature vector. Specifically, in the embodiment of the present application, the sequence of the local temporal correlation feature vectors of the heart rate is subjected to the local temporal correlation feature extraction of the heart rate to obtain the first time scale heart rate pattern feature vector, including: passing the sequence of the local temporal correlation feature vectors of the heart rate through the local temporal correlation feature extractor of the heart rate of the one-dimensional extended convolutional neural network model to obtain the first time scale heart rate pattern feature vector. Accordingly, considering that each of the local temporal correlation feature vectors of the heart rate in the sequence of the local temporal correlation feature vectors of the heart rate has a mutual correlation relationship with respect to the local temporal characteristics of the heart rate, that is, the local temporal correlation feature of the heart rate at a certain moment is usually related to the local temporal correlation feature of the heart rate at a nearby moment, and there is also semantic correlation information about the heart rate data between each of the local temporal correlation feature vectors of different time lengths in the local temporal correlation feature vectors of the heart rate. Therefore, in order to focus on the semantic inter-correlation of the heart rate data at different time steps, in the technical solution of the present application, the sequence of the heart rate local temporal correlation feature vectors is passed through the heart rate local temporal domain correlation feature extractor of the one-dimensional extended convolutional neural network model to obtain the first time scale heart rate pattern feature vector. It should be understood that by capturing and correlating the heart rate local temporal correlation features at different time lengths in the sequence of the heart rate local temporal correlation feature vectors through the one-dimensional extended convolutional neural network model, it is possible to extract higher-level heart rate time domain semantic correlation features, thereby improving the understanding and expression capabilities of the heart rate temporal pattern, so as to obtain a more representative first time scale heart rate pattern feature vector.

[0048] More specifically, in an embodiment of the present application, the sequence of the heart rate local time series correlation feature vectors is passed through the heart rate local time domain correlation feature extractor of the one-dimensional extended convolutional neural network model to obtain the first time scale heart rate pattern feature vector, including: passing the sequence of the heart rate local time series correlation feature vectors through the heart rate local time domain correlation feature extractor of the one-dimensional extended convolutional neural network model with the following extended convolution formula to process to obtain the first time scale heart rate pattern feature vector; wherein the extended convolution formula is:

[0049] X 1:n =X 1 ⊕X 2 ⊕...⊕X n

[0050] C i =f(ω·X i:i+l-1 +b)

[0051] C=[C 1 ,C 2 ,...,C n-l+1 ]

[0052] Among them, X 1 ,X 2 ,...,X n represents each heart rate local time series associated feature vector in the sequence of the heart rate local time series associated feature vector, X 1:n represents the concatenated vector of the sequence of the local temporal correlation feature vectors of the heart rate, ⊕ represents the connection operation, X i:i+l-1 X represents the sequence of the local temporal correlation feature vector of the heart rate i ,X i+1 ,...,X i+l-1 , ω and b represent the weight matrix and bias vector respectively, f(·) represents the convolution operation, C i represents the i-th heart rate local temporal enhancement association feature vector, and C represents the heart rate pattern feature vector of the first time scale.

[0053] In step S160, the sequence of the local temporal association feature vectors of the heart rate is subjected to the global temporal association feature extraction of the heart rate to obtain the second time scale heart rate pattern feature vector. Specifically, in an embodiment of the present application, the sequence of the local temporal association feature vectors of the heart rate is subjected to the global temporal association feature extraction of the heart rate to obtain the second time scale heart rate pattern feature vector, including: passing the sequence of the local temporal association feature vectors of the heart rate through a global temporal association feature extractor of the heart rate based on a converter module to obtain the second time scale heart rate pattern feature vector. It should be understood that, considering that each local temporal association feature vector of the heart rate in the sequence of the local temporal association feature vectors of the heart rate has a semantic association relationship between contexts on a time scale. Therefore, in order to more accurately understand and analyze the semantic association feature information between each local temporal association feature of the heart rate in the sequence of the local temporal association feature vectors of the heart rate, in the technical solution of the present application, the sequence of the local temporal association feature vectors of the heart rate is passed through a global temporal association feature extractor of the heart rate based on a converter module to obtain the second time scale heart rate pattern feature vector. Specifically, the transformer module is able to capture long-range dependencies in sequence data through the self-attention mechanism, which is crucial for understanding global patterns and trends in heart rate data. That is, the self-attention mechanism in the transformer module allows the model to consider the local temporal semantic associations of heart rate between all other points in the sequence of the local temporal association feature vector of heart rate at each time point, thereby capturing the second time scale heart rate pattern feature vector containing global temporal association semantic information.

[0054] In step S170, the first time scale heart rate pattern feature vector and the second time scale heart rate pattern feature vector are fused using a dynamic element-level feature fusion module to obtain a multi-scale heart rate pattern feature vector as a multi-scale heart rate pattern feature. Accordingly, considering that the first time scale heart rate pattern feature vector and the second time scale heart rate pattern feature vector respectively reflect the local temporal correlation features of heart rate at different scales, in order to effectively integrate the local temporal correlation features of heart rate from different time scales, thereby providing a comprehensive heart rate pattern feature representation, in the technical solution of the present application, the first time scale heart rate pattern feature vector and the second time scale heart rate pattern feature vector are fused using the dynamic element-level feature fusion module to obtain the multi-scale heart rate pattern feature vector as a multi-scale heart rate pattern feature. It is worth mentioning that the dynamic element-level feature fusion module adaptively balances the importance of the fusion information from the input vector by introducing a threshold value, and dynamically allocates weights according to the importance of the features to automatically adjust the proportion of feature fusion, thereby improving the efficiency of feature utilization. That is to say, by using the dynamic element-level feature fusion module to fuse the first time-scale heart rate pattern feature vector and the second time-scale heart rate pattern feature vector, it is possible to more finely process the important and key heart rate features in the heart rate pattern, thereby helping the system to better understand the evolution process and trend changes of the heart rate pattern, making the obtained multi-scale heart rate pattern feature vector more representative and distinctive, and providing richer and more reliable data support for subsequent heart rate abnormality assessment.

[0055] Specifically, in an embodiment of the present application, the first time scale heart rate pattern feature vector and the second time scale heart rate pattern feature vector are fused using the dynamic element level feature fusion module to obtain the multi-scale heart rate pattern feature vector as a multi-scale heart rate pattern feature, including: using the dynamic element level feature fusion module to fuse the first time scale heart rate pattern feature vector and the second time scale heart rate pattern feature vector using the following fusion formula to obtain the multi-scale heart rate pattern feature vector; wherein the fusion formula is:

[0056]

[0057] in, and are the eigenvalues ​​of each position in the first time scale heart rate pattern eigenvector and the second time scale heart rate pattern eigenvector, t is the threshold value, t∈[0,1], v g is the eigenvalue of each position in the multi-scale heart rate pattern eigenvector, V 1 g and V 2 gare the first time scale heart rate pattern feature vector and the second time scale heart rate pattern feature vector respectively, T is the threshold value of each position in the threshold vector, W 0 is the transformation matrix, b is the bias vector, sigmoid(·) represents the activation function, and concat(·,·) represents the cascade operation.

[0058] In step S180, based on the multi-scale heart rate pattern features, an early warning result is obtained. Specifically, in an embodiment of the present application, based on the multi-scale heart rate pattern features, an early warning result is obtained, including: passing the multi-scale heart rate pattern feature vector through a classifier-based heart rate early warning device to obtain an early warning result, and the early warning result is used to indicate whether an abnormal heart rate early warning prompt is generated. That is, the multi-scale heart rate pattern features obtained by dynamic element-level feature fusion of the first time scale heart rate pattern feature vector and the second time scale heart rate pattern feature vector are classified and processed, so as to intelligently determine whether an abnormal heart rate early warning prompt is generated according to the multi-scale heart rate pattern features of the heart rate data of the monitored student object. In this way, the low power consumption and efficient transmission characteristics of BLE technology can be used to ensure the real-time and accuracy of the data. At the same time, the system can quickly and accurately identify abnormal heart rate patterns and issue early warning prompts when necessary, reduce false alarms and missed alarms, and thus provide timely intervention and support for students' health management.

[0059] In the technical solution described above, the first time scale heart rate pattern feature vector and the second time scale heart rate pattern feature vector respectively express the local temporal correlation characteristics of the time series of the heart rate data in local time domains of different scales divided by different scales in the global time domain. Thus, the first time scale heart rate pattern feature vector and the second time scale heart rate pattern feature vector have the distinctiveness of temporal correlation feature expressions corresponding to different temporal correlation patterns in local time domains of different scales.

[0060] Specifically, when using the dynamic element-level feature fusion module to process the first time-scale heart rate pattern feature vector and the second time-scale heart rate pattern feature vector to obtain the multi-scale heart rate pattern feature vector, the dynamic element-level feature fusion offset will occur due to the difference in the temporal feature receptive fields of the first time-scale heart rate pattern feature vector and the second time-scale heart rate pattern feature vector. It is expected that the comprehensibility of the multi-scale temporal dynamic fusion of the multi-scale heart rate pattern feature vector will be further improved, thereby improving the accuracy of the warning results obtained by the heart rate warning device based on the classifier as input.

[0061] Therefore, when the multi-scale heart rate pattern feature vector is passed through a classifier-based heart rate alarm to obtain an early warning result, the present application optimizes the multi-scale heart rate pattern feature vector, including the steps of: subtracting the 0 norm of the multi-scale heart rate pattern feature vector from the length of the multi-scale heart rate pattern feature vector to obtain a multi-scale heart rate pattern isolated representation value; calculating the logarithm of the square of the multi-scale heart rate pattern isolated representation value and the sum of the multi-scale heart rate pattern isolated representation value with base 2 to obtain a multi-scale heart rate pattern information order value; calculating the difference between the multi-scale heart rate pattern isolated representation value minus one and each eigenvalue of the multi-scale heart rate pattern feature vector as the base. The power function of the number is multiplied by the multi-scale heart rate pattern information order value to obtain a multi-scale heart rate pattern guide vector; the multi-scale heart rate pattern feature vector is multiplied by the difference between the multi-scale heart rate pattern isolated representation value minus one, and then multiplied by the reciprocal of the multi-scale heart rate pattern isolated representation value to obtain a multi-scale heart rate pattern field constraint vector; the exponential function of each eigenvalue of the multi-scale heart rate pattern field constraint vector is calculated with the natural constant as the base to obtain a multi-scale heart rate pattern field bias vector; the multi-scale heart rate pattern guide vector is multiplied by the multi-scale heart rate pattern field bias vector to obtain an optimized multi-scale heart rate pattern feature vector. Finally, the optimized multi-scale heart rate pattern feature vector is passed through a classifier-based heart rate early warning device to obtain the early warning result.

[0062] The optimization process of the multi-scale heart rate pattern feature vector, for example, denoted as V, is expressed as:

[0063] V′=V ⊙(n-1) ⊙log 2 (n 2 +n)⊕exp[V⊙(n-1)⊙n -1 ]

[0064] n=L-‖V‖ :

[0065] Where V represents the multi-scale heart rate pattern feature vector, ‖·‖ : represents the zero norm of the vector, L represents the length of the multi-scale heart rate pattern feature vector, n represents the isolated representation value of the multi-scale heart rate pattern, ⊙ represents the point multiplication by position, ⊕ represents the addition by position, (·) ⊙(n-1) represents a power function with the difference value of the multi-scale heart rate pattern isolated representation value minus one as the exponent for calculating the feature value of each position in the feature vector, and V′ represents the optimized multi-scale heart rate pattern feature vector.

[0066] Therefore, the high-dimensional feature manifold of the multi-scale heart rate pattern feature vector is represented by the eigenvalues ​​of the feature set as the vector field of the aggregation dimension, and the superposition value of the vector field of the multi-scale heart rate pattern feature vector at the isolated zero position is used as the order information to fix the local position of the eigenvalue of its feature set, and a bias for the reversibility of the feature regression distribution field of the multi-scale heart rate pattern feature vector is added as a reward to achieve the mapping target tracking of the regression distribution of the multi-scale heart rate pattern feature vector for the eigenvalue position, so that the feature set of the multi-scale heart rate pattern feature vector can perceive the mapping migration to the aggregate distribution, thereby improving the accuracy of the warning result obtained by the heart rate warning device based on the classifier when the multi-scale heart rate pattern feature vector is input into the multi-scale heart rate pattern feature vector by improving the comprehensibility of the multi-scale time series dynamic fusion of the multi-scale heart rate pattern feature vector.

[0067] In summary, the physical fitness data collection method based on BLE technology according to the embodiment of the present application is explained, which can collect the heart rate data of the monitored student object in real time through the smart wearable device, and transmit the heart rate data to the cloud platform using BLE technology, so as to use the artificial intelligence-based data processing and analysis algorithm in the cloud platform to perform time series feature analysis and multi-scale association on the heart rate data of the monitored student object, so as to intelligently determine whether to generate an abnormal heart rate warning prompt based on the multi-scale heart rate pattern characteristics of the heart rate data of the monitored student object. In this way, the low power consumption and efficient transmission characteristics of BLE technology can be used to ensure the real-time and accuracy of the data. At the same time, the system can quickly and accurately identify abnormal heart rate patterns and issue warning prompts when necessary, reduce false alarms and missed alarms, and thus provide timely intervention and support for students' health management.

[0068] Figure 4 FIG. 1 is a block diagram of a physical fitness data collection system based on BLE technology according to an embodiment of the present application. Figure 4As shown, according to the embodiment of the present application, the physical fitness data acquisition system 100 based on BLE technology includes: a heart rate data acquisition module 110, which is used to collect the time series of heart rate data of the monitored student object through a smart wearable device; a heart rate data uploading module 120, which is used to upload the time series of the heart rate data to the cloud platform through the BEL module; a heart rate data timing arrangement module 130, which is used to arrange the time series of the heart rate data into a heart rate timing input vector according to the time dimension on the cloud platform; a vector segmentation feature extraction module 140, which is used to perform vector segmentation on the heart rate timing input vector on the cloud platform to obtain a sequence of heart rate local timing input vectors, and then perform heart rate timing pattern feature extraction on the sequence of the heart rate local timing input vectors to obtain heart rate local timing related A sequence of related feature vectors; a local time-domain correlation module 150, used to extract the local time-domain correlation features of the heart rate on the sequence of local time-series correlation feature vectors to obtain a first time-scale heart rate pattern feature vector; a global time-domain correlation module 160, used to extract the global time-domain correlation features of the heart rate on the sequence of local time-series correlation feature vectors to obtain a second time-scale heart rate pattern feature vector; a feature fusion module 170, used to use a dynamic element-level feature fusion module to fuse the first time-scale heart rate pattern feature vector and the second time-scale heart rate pattern feature vector to obtain a multi-scale heart rate pattern feature vector as a multi-scale heart rate pattern feature; and an early warning result generation module 180, used to obtain an early warning result based on the multi-scale heart rate pattern feature.

[0069] Specifically, the specific operations of each step in the above-mentioned physical fitness data collection system based on BLE technology have been referred to above. Figures 1 to 3 The description of the physical fitness data collection method based on BLE technology has been introduced in detail, and therefore, its repeated description will be omitted.

[0070] As described above, the physical fitness data collection system 100 based on BLE technology according to the embodiment of the present disclosure can be implemented in various wireless terminals, such as a server with a physical fitness data collection algorithm based on BLE technology. In one possible implementation, the physical fitness data collection system 100 based on BLE technology according to the embodiment of the present disclosure can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the physical fitness data collection system 100 based on BLE technology can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the physical fitness data collection system 100 based on BLE technology can also be one of the many hardware modules of the wireless terminal.

[0071] Alternatively, in another example, the physical fitness data acquisition system 100 based on BLE technology and the wireless terminal may also be separate devices, and the physical fitness data acquisition system 100 based on BLE technology may be connected to the wireless terminal through a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0072] The foregoing is merely an example of the principles of the present disclosure, and various modifications may be made by those skilled in the art without departing from the scope of the present disclosure. The above embodiments are presented for the purpose of illustration and not limitation. The present disclosure may also take many forms other than those explicitly described herein. Therefore, it is emphasized that the present disclosure is not limited to the methods, systems and devices explicitly disclosed, but is intended to include variations and modifications within the spirit and scope of the appended claims.

Claims

1. A physical data collection method based on BLE technology, It is characterized in that include: Collect the time series of heart rate data of monitored student subjects through smart wearable devices; Upload the time series of the heart rate data to the cloud platform through the BEL module; On the cloud platform, arranging the time series of the heart rate data into a heart rate time series input vector according to the time dimension; On the cloud platform, performing vector segmentation on the heart rate timing input vector to obtain a sequence of heart rate local timing input vectors, and then performing heart rate timing pattern feature extraction on the sequence of heart rate local timing input vectors to obtain a sequence of heart rate local timing associated feature vectors; Extracting local time domain correlation features of the heart rate from the sequence of local time series correlation feature vectors of the heart rate to obtain a first time scale heart rate pattern feature vector; Extracting the global time domain correlation features of the heart rate from the sequence of the local time series correlation feature vectors of the heart rate to obtain a second time scale heart rate pattern feature vector; Using a dynamic element-level feature fusion module to fuse the first time-scale heart rate pattern feature vector and the second time-scale heart rate pattern feature vector to obtain a multi-scale heart rate pattern feature vector as a multi-scale heart rate pattern feature; Based on the multi-scale heart rate pattern characteristics, an early warning result is obtained.

2. According to the physical fitness data collection method based on BLE technology according to claim 1, It is characterized in that On the cloud platform, vector segmentation is performed on the heart rate timing input vector to obtain a sequence of heart rate local timing input vectors, and then heart rate timing pattern feature extraction is performed on the sequence of heart rate local timing input vectors to obtain a sequence of heart rate local timing associated feature vectors, including: On the cloud platform, vector segmentation is performed on the heart rate time series input vector based on a predetermined time scale to obtain a sequence of the heart rate local time series input vectors; On the cloud platform, the sequence of the heart rate local timing input vectors is passed through a heart rate timing pattern feature extractor based on a one-dimensional convolutional layer to obtain a sequence of the heart rate local timing associated feature vectors.

3. According to the physical fitness data collection method based on BLE technology according to claim 2, It is characterized in that The sequence of the heart rate local time series correlation feature vectors is subjected to heart rate local time domain correlation feature extraction to obtain a first time scale heart rate pattern feature vector, comprising: passing the sequence of the heart rate local time series correlation feature vectors through a heart rate local time domain correlation feature extractor of a one-dimensional extended convolutional neural network model to obtain the first time scale heart rate pattern feature vector.

4. According to the physical fitness data collection method based on BLE technology according to claim 3, It is characterized in that The sequence of the local temporal correlation feature vectors of the heart rate is passed through the local temporal correlation feature extractor of the heart rate of the one-dimensional extended convolutional neural network model to obtain the first time scale heart rate pattern feature vector, including: the sequence of the local temporal correlation feature vectors of the heart rate is passed through the local temporal correlation feature extractor of the heart rate of the one-dimensional extended convolutional neural network model to process with the following extended convolution formula to obtain the first time scale heart rate pattern feature vector; Wherein, the extended convolution formula is: C i =f(ω·X i:i+l-1 +b) C=[C 1 ,C 2 ,...,C n-l+1 ] Among them, X 1 ,X 2 ,...,X n represents each heart rate local time series associated feature vector in the sequence of the heart rate local time series associated feature vector, X 1:n A concatenated vector representing a sequence of local temporal correlation feature vectors of the heart rate, Indicates the connection operation, X i:i+l-1 X represents the sequence of the local temporal correlation feature vector of the heart rate i ,X i+1 ,...,X i+l-1 , ω and b represent the weight matrix and bias vector respectively, f(·) represents the convolution operation, C i represents the i-th heart rate local temporal enhancement association feature vector, and C represents the heart rate pattern feature vector of the first time scale.

5. According to the physical fitness data collection method based on BLE technology according to claim 4, It is characterized in that The sequence of the heart rate local time series correlation feature vectors is subjected to heart rate global time domain correlation feature extraction to obtain a second time scale heart rate pattern feature vector, comprising: passing the sequence of the heart rate local time series correlation feature vectors through a heart rate global time domain correlation feature extractor based on a converter module to obtain the second time scale heart rate pattern feature vector.

6. The method for collecting physical fitness data based on BLE technology according to claim 5, It is characterized in that Using the dynamic element-level feature fusion module to fuse the first time-scale heart rate pattern feature vector and the second time-scale heart rate pattern feature vector to obtain the multi-scale heart rate pattern feature vector as a multi-scale heart rate pattern feature, including: using the dynamic element-level feature fusion module to fuse the first time-scale heart rate pattern feature vector and the second time-scale heart rate pattern feature vector using the following fusion formula to obtain the multi-scale heart rate pattern feature vector; Among them, the fusion formula is: in, and are the eigenvalues ​​of each position in the first time scale heart rate pattern eigenvector and the second time scale heart rate pattern eigenvector, t is the threshold value, t∈[0,1], v g is the eigenvalue of each position in the multi-scale heart rate pattern eigenvector, V 1 g and are the first time scale heart rate pattern feature vector and the second time scale heart rate pattern feature vector respectively, T is the threshold value of each position in the threshold vector, W 0 is the transformation matrix, b is the bias vector, sigmoid(·) represents the activation function, and concat(·,·) represents the cascade operation.

7. The method for collecting physical fitness data based on BLE technology according to claim 6, It is characterized in that Based on the multi-scale heart rate pattern features, an early warning result is obtained, including: the multi-scale heart rate pattern feature vector is passed through a classifier-based heart rate early warning device to obtain an early warning result, and the early warning result is used to indicate whether an abnormal heart rate early warning prompt is generated.

8. A physical data collection system based on BLE technology, It is characterized in that include: A heart rate data acquisition module is used to collect the time series of heart rate data of the monitored student object through a smart wearable device; A heart rate data uploading module, used to upload the time series of the heart rate data to the cloud platform through the BEL module; A heart rate data time series arrangement module, used for arranging the time series of the heart rate data into a heart rate time series input vector according to the time dimension on the cloud platform; A vector segmentation feature extraction module is used to perform vector segmentation on the heart rate timing input vector on the cloud platform to obtain a sequence of heart rate local timing input vectors, and then perform heart rate timing pattern feature extraction on the sequence of heart rate local timing input vectors to obtain a sequence of heart rate local timing associated feature vectors; A local time domain correlation module, used for extracting the local time domain correlation features of the heart rate from the sequence of the local time series correlation feature vectors of the heart rate to obtain a first time scale heart rate pattern feature vector; A global time domain association module, used for extracting the global time domain association features of the heart rate from the sequence of the local time series association feature vectors of the heart rate to obtain a second time scale heart rate pattern feature vector; A feature fusion module is used to use a dynamic element level feature fusion module to fuse the first time scale heart rate pattern feature vector and the second time scale heart rate pattern feature vector to obtain a multi-scale heart rate pattern feature vector as a multi-scale heart rate pattern feature.

9. The physical fitness data collection system based on BLE technology according to claim 8, It is characterized in that The global time domain association module is used to: pass the sequence of the heart rate local time series association feature vectors through a heart rate global time domain association feature extractor based on a converter module to obtain the second time scale heart rate pattern feature vector.

10. The physical fitness data collection system based on BLE technology according to claim 8, It is characterized in that It also includes a warning result generation module, which is used to obtain a warning result based on the multi-scale heart rate pattern characteristics.

Images