Pericardial health data analysis and processing system, electronic devices, and readable storage media

Abstract

This application provides a pericardial health data analysis and processing system, electronic device, and readable storage medium, relating to the field of medical and health data processing technology. In the system, first physiological data and second physiological data collected by a wearable device are received. The first physiological data includes raw PPG waveform signals. Feature processing is performed on the raw PPG waveform signals to obtain pericardial pulse characteristics. Dynamic baseline data is acquired, obtained by processing historical physiological data collected by the wearable device from the user through a preset dynamic baseline analysis model. The pericardial pulse characteristics, second physiological data, and dynamic baseline data are fused to obtain a multi-dimensional feature vector. The multi-dimensional feature vector is input into a preset multimodal physiological state assessment model for processing to obtain analysis and assessment results and traditional Chinese medicine (TCM) characteristic state labels. Based on the analysis and assessment results and TCM characteristic state labels, a pericardial data analysis report is generated, realizing intelligent analysis of pericardial health risks.

Description

Technical Field

[0001] This application relates to the field of medical and health data processing technology, specifically to a pericardial health data analysis and processing system, electronic device, and readable storage medium. Background Technology

[0002] With advancements in sensor technology, modern smart wearable devices can now collect physiological data such as heart rate (HR), blood pressure, blood oxygen (SpO2), and heart rate variability (HRV) with high precision, demonstrating excellent performance in detecting sudden cardiovascular events. However, current methods for analyzing physiological data collected by wearable devices are still primarily based on Western medical anatomy and electrophysiology frameworks. These methods typically process data uniformly across all time periods, triggering an alert when a certain physiological indicator exceeds a preset threshold. This analysis process cannot accurately monitor the overall blood and qi status of the body. For example, many users who experience chest tightness, palpitations, or restlessness go to the hospital for an electrocardiogram or wear a fitness tracker, only to find that the results show "normal indicators." Relying solely on data collected by smart wearable devices and employing conventional, uniform sampling strategies makes it difficult to capture subtle changes in characteristics at specific times of day, thus hindering intelligent health management. Summary of the Invention

[0003] To address the aforementioned issues, this application provides a pericardial health data analysis and processing system, electronic device, and readable storage medium, which enables intelligent integration of wearable devices and traditional Chinese medicine theory, captures subtle feature changes at specific times, and achieves intelligent analysis of pericardial health risks.

[0004] The technical solution of this application embodiment is as follows: In a first aspect, embodiments of this application provide a pericardial health data analysis and processing system. The system is communicatively connected to a wearable device. The wearable device is used to collect first physiological data of a user according to a preset first rule and second physiological data of the user according to a second rule. The first rule involves sampling at a first frequency during a first time period, and the second rule involves sampling at a second frequency during a second time period. The first time period corresponds to a physiologically active period, and the second time period is a period of the day excluding the first time period. The first frequency is greater than the second frequency. The system includes: The data receiving module is used to receive the first physiological data and the second physiological data collected by the wearable device, wherein the first physiological data includes the PPG raw waveform signal; The feature processing module is used to perform feature processing on the original PPG waveform signal to obtain pericardial pulse characteristics; A data acquisition module, configured to acquire dynamic baseline data, which is obtained by processing the historical physiological data of a user collected by the wearable device through a preset dynamic baseline analysis model; A data fusion module, configured to fuse the pericardium pulse characteristics, the second physiological data, and the dynamic baseline data to obtain a multi-dimensional feature vector; An evaluation and report generation module, configured to input the multi-dimensional feature vector into a preset multi-modal physiological state evaluation model for processing to obtain an analysis and evaluation result and a traditional Chinese medicine feature state label, and generate a pericardium data analysis report based on the analysis and evaluation result and the traditional Chinese medicine feature state label.

[0005] In the above technical solution, the analysis and processing system is communicatively connected to the wearable device so as to subsequently acquire the data collected by the wearable device for processing, implement relevant data analysis, and achieve intelligent analysis of pericardium health risks. The wearable device is configured to collect the first physiological data of the user according to a preset first rule and collect the second physiological data of the user according to a second rule. The first rule is to sample at a first frequency during a first time period, and the second rule is to sample at a second frequency during a second time period. The first time period corresponds to a physiologically active time period, and the second time period is the time period other than the first time period on the same day. The first frequency is greater than the second frequency. Different from the data collection of existing wearable devices, the first physiological data of this time period is sampled at a high frequency of the first frequency according to traditional Chinese medicine's Wu Shi to ensure that sufficient fine waveform data (reflecting the flow of qi and blood in the pericardium meridian in traditional Chinese medicine theory) is obtained, and the second physiological data of the whole day is sampled at a low frequency of the second frequency during the second time period, so as to achieve comprehensive data sampling and provide a data basis for subsequent calculations.

[0006] Based on the above settings, the first physiological data and the second physiological data collected by the wearable device are received. The first physiological data includes the PPG raw waveform signal, so as to process the PPG raw waveform signal subsequently and provide data support; perform feature processing on the PPG raw waveform signal to obtain pericardial pulse characteristics, convert the collected waveform signal into characteristics that can reflect traditional Chinese medicine theory, which represent abnormal states in traditional Chinese medicine theory, and also achieve the intelligent combination of mapping the data collected by the wearable device to traditional Chinese medicine theory; obtain dynamic baseline data, which is obtained by processing the historical physiological data of the user collected by the wearable device through a preset dynamic baseline analysis model. By obtaining dynamic baseline data, other parameters of the human body are reflected, comprehensive data is obtained, and the accuracy of analysis and evaluation is achieved; fuse the pericardial pulse characteristics, the second physiological data and the dynamic baseline data to obtain a multi-dimensional feature vector. Through data fusion, the comprehensiveness and expression ability of data processing can be increased, providing support for accurate risk analysis; input the multi-dimensional feature vector into a preset multi-modal physiological state evaluation model for processing to obtain an analysis and evaluation result and a traditional Chinese medicine feature state label, and generate a pericardial data analysis report based on the analysis and evaluation result and the traditional Chinese medicine feature state label, realizing intelligent health management.

[0007] In some embodiments of the present application, the period of physiological activity is the xushi in traditional Chinese medicine. The feature processing module includes a first feature analysis sub-module, a second feature analysis sub-module, a third feature analysis sub-module and a feature fusion sub-module. The first feature analysis sub-module is used to analyze the pulse position and pulse potential of the PPG raw waveform signal to obtain pulse position and pulse potential characteristics. The second feature analysis sub-module is used to analyze the pulse shape of the PPG raw waveform signal to obtain pulse shape characteristics. The third feature analysis sub-module is used to analyze the pulse rate of the PPG raw waveform signal to obtain pulse rate characteristics. The feature fusion sub-module is used to perform vector fusion processing on the pulse position and pulse potential characteristics, the pulse shape characteristics and the pulse rate characteristics to obtain the pericardial pulse characteristics.

[0008] In some embodiments of the present application, the second feature analysis sub-module includes a stiffness calculation sub-module, a rhythm statistics sub-module, a smoothness calculation sub-module and a calculation fusion sub-module. The stiffness calculation sub-module is used to calculate the waveform stiffness index based on the slope of the rising branch of the waveform of the PPG raw waveform signal and the ratio of height to the time difference of the dicrotic wave, and perform quantization to obtain the vascular wall tension. The rhythm statistics submodule is used to perform regularity statistics on the waveform period of the PPG raw waveform signal to obtain irregular rhythms, and to quantify the irregular rhythms to obtain the pulse degree. The smoothness calculation submodule is used to calculate the smoothness of the falling branch of the PPG original waveform signal and quantize it to obtain the blood flow smoothness. The computational fusion submodule is used to perform vector fusion processing on the vessel wall tension, the pulse pause degree, and the blood flow smoothness to obtain the pulse shape feature.

[0009] In some embodiments of this application, the data fusion module includes a weighted fusion submodule, an association calculation submodule, a search submodule, and a vector construction submodule. The weighted fusion submodule is used to perform weighted fusion of the pericardial pulse features and the second physiological data to obtain a first-dimensional feature vector, wherein the weight of the pericardial pulse features is greater than the weight of the second physiological data. The association calculation submodule is used to associate the pericardial pulse characteristics with the vital sign data and status data in the dynamic baseline data to obtain a second-dimensional feature vector. The search submodule is used to search for index features that are associated with the pericardial pulse features in a preset meridian association table, and obtain a third-dimensional feature vector. The meridian association table includes the correspondence between the pericardial pulse features and preset triple energizer meridian indicators and liver meridian indicators. The vector construction submodule is used to construct the multi-dimensional feature vector from the first-dimensional feature vector, the second-dimensional feature vector, and the third-dimensional feature vector.

[0010] In some embodiments of this application, the correlation calculation submodule includes a feature extraction submodule, an anomaly analysis submodule, and a feature concatenation submodule. The feature extraction submodule is used to extract heart rate signs and emotional state from the vital signs data that are in the same period as the pericardial pulse characteristics. The anomaly analysis submodule is used to perform anomaly analysis on the heart rate signs and obtain anomaly analysis results; The feature splicing submodule is used to determine the confidence level of the anomaly analysis result using the emotional state. If the confidence level is greater than a preset confidence threshold, the vital signs data, the emotional state, and the pericardial pulse characteristics are spliced ​​together to obtain the second-dimensional feature vector.

[0011] In some embodiments of this application, the evaluation and report generation module includes a label mapping submodule and a report generation submodule. The label mapping submodule is used to map the TCM characteristic status label to a preset warning prompt when the analysis and evaluation result is greater than a preset risk threshold, so as to obtain status prompt information. The report generation submodule is used to generate the pericardial data analysis report by combining the TCM characteristic status labels and the status prompt information.

[0012] In some embodiments of this application, the analysis and processing system further includes a suggestion query module and a real-time tracking module. The suggestion query module is used to search for conditioning suggestion information from a preset conditioning plan library based on the pericardial data analysis report after the pericardial data analysis report is generated based on the analysis and evaluation results and the TCM characteristic status labels. The real-time tracking module is used to push the pericardial data analysis report and the treatment suggestion information to the client, and receive the treatment effect feedback from the client for real-time tracking.

[0013] Secondly, embodiments of this application provide an electronic device, including a processor, a memory, a user interface, a communication bus, and a network interface. The processor, the memory, the user interface, and the network interface are respectively connected to the communication bus. The memory is used to store instructions. The user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the steps in any of the systems provided in the first aspect.

[0014] Thirdly, embodiments of this application provide a computer-readable storage medium storing instructions that, when executed, perform the steps in any of the systems provided in the first aspect above.

[0015] In summary, one or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. Firstly, the data collection of wearable devices differs from that of existing wearable devices. During physiologically active periods, the first physiological data of that period is sampled at a high frequency to ensure that sufficiently detailed waveform data is obtained (reflecting the flow of Qi and blood in the pericardium meridian in traditional Chinese medicine theory). In the second period, the second physiological data of the user for the remaining time of the day is sampled at a low frequency to achieve comprehensive data sampling and provide a data foundation for subsequent calculations. The system receives first and second physiological data collected by a wearable device. The first physiological data includes raw PPG waveform signals for subsequent processing to provide data support. Feature processing is performed on the raw PPG waveform signals to obtain pericardial pulse characteristics, converting the collected waveform signals into features that reflect traditional Chinese medicine (TCM) theory. These features represent abnormal states according to TCM theory, and the system also achieves an intelligent mapping of wearable device data to TCM theory. Dynamic baseline data is acquired by processing historical physiological data collected by the wearable device using a preset dynamic baseline analysis model. This dynamic baseline data reflects other parameters of the human body, providing comprehensive data and improving the accuracy of analysis and assessment. The pericardial pulse characteristics, second physiological data, and dynamic baseline data are fused to obtain a multi-dimensional feature vector. This data fusion increases the comprehensiveness and expressiveness of data processing, supporting accurate risk analysis. The multi-dimensional feature vector is then input into a preset multimodal physiological state assessment model for processing, yielding analysis and assessment results and TCM characteristic state labels. Based on these results and labels, a pericardial data analysis report is generated, enabling intelligent health management. Therefore, it effectively solves the problem in related technologies that it is difficult to capture subtle feature changes at specific times using conventional uniform sampling strategies based on data collected from smart wearable devices, thus failing to achieve intelligent health management.

[0016] 2. Transforming the traditionally tactile pericardial pulse characteristics (wiry, hurried, intermittent, and hesitant) into quantifiable PPG waveform feature parameters marks a crucial step towards objective and standardized signal representation in TCM data analysis.

[0017] 3. By combining dynamic baseline analysis models and multimodal physiological state assessment models, analysis results can be provided and early warnings can be issued before users show obvious serious abnormalities. Treatment suggestions can be pushed to the client, providing highly personalized TCM health preservation guidance and realizing intelligent health management.

[0018] 4. It has connected the entire chain from hardware signal acquisition, artificial intelligence analysis, TCM theory feature mapping and health management services, forming a complete technology closed loop, avoiding data redundancy and power consumption balance. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the processing structure of a pericardial health data analysis and processing system provided in one embodiment of this application; Figure 2 This is a schematic diagram of the correlation calculation submodule of the pericardial health data analysis and processing system provided in another embodiment of this application; Figure 3 This is a schematic diagram of a multimodal physiological state assessment model of a pericardial health data analysis and processing system provided in one embodiment of this application; Figure 4 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0021] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0022] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0023] In the related art, traditional Chinese medicine can reflect the overall qi and blood state of the human body. In traditional Chinese medicine theory, the "pericardium" is regarded as the "palace city of the heart" and is responsible for receiving pathogenic factors on behalf of the heart. Abnormal functions of the pericardium meridian are closely related to symptoms such as "palpitation", "chest distress", "restlessness of mind", as well as some arrhythmias, cardiac neurosis, and even early manifestations of coronary heart disease in modern medicine. The traditional assessment of the function of the pericardium meridian mainly relies on Chinese medicine practitioners' "observation, auscultation and olfaction, interrogation, and palpation", especially pulse palpation, which is highly subjective and difficult to achieve continuous and quantitative dynamic monitoring. That is, the collected data cannot be quantified and feature-processed for intuitive data display. In the related art, the detected physiological data cannot be correlated and mapped with the qi and blood flow rhythm of the Chinese medicine meridian to achieve intelligent analysis and effective health management. Moreover, there is a lack of early and quantitative assessment methods based on multi-dimensional data fusion for "pericardium" data.

[0024] Based on this, the embodiments of the present application provide an analysis and processing system for pericardial health data, an electronic device, and a readable storage medium. The analysis and processing system for pericardial health data is communicatively connected to a wearable device to subsequently obtain the data collected by the wearable device for processing, realize relevant data analysis, and achieve intelligent analysis of pericardial health risks. The wearable device is used to collect the first physiological data of the user according to a preset first rule and the second physiological data of the user according to a second rule. Among them, the first rule is to sample at a first frequency during a first time period, and the second rule is to sample at a second frequency during a second time period. The first time period corresponds to the period of physiological activity, and the second time period is the period of the day other than the first time period. The first frequency is greater than the second frequency. Different from the data collection of wearable devices in the prior art, the first physiological data of this time period is sampled at a high frequency according to the first frequency at the xushi time in traditional Chinese medicine to ensure obtaining sufficiently fine waveform data (reflecting the qi and blood flow of the pericardium meridian in traditional Chinese medicine theory), and the second physiological data of the whole day is sampled at a low frequency according to the second frequency during the second time period, so as to achieve comprehensive data sampling and provide a data basis for subsequent calculations.

[0025] Refer to Figure 1The pericardial health data analysis and processing system 100 receives first and second physiological data collected by a wearable device through a data receiving module 110. The first physiological data includes raw PPG waveform signals for subsequent processing to provide data support. Then, the feature processing module 120 performs feature processing on the raw PPG waveform signals to obtain pericardial pulse characteristics, converting the collected waveform signals into features that reflect traditional Chinese medicine theory. These features represent abnormal states in traditional Chinese medicine theory and also realize the intelligent combination of mapping data collected by the wearable device to traditional Chinese medicine theory. The data acquisition module 130 acquires dynamic baseline data, which is then analyzed by a preset dynamic baseline analysis model on the wearable device. The system collects and processes users' historical physiological data. By acquiring dynamic baseline data, it reflects other parameters of the human body, obtaining comprehensive data to ensure accurate analysis and assessment. The data fusion module 140 fuses pericardial pulse characteristics, secondary physiological data, and dynamic baseline data to obtain multi-dimensional feature vectors. Data fusion increases the comprehensiveness and expressiveness of data processing, providing support for accurate risk analysis. Finally, the assessment and report generation module 150 inputs the multi-dimensional feature vectors into a preset multimodal physiological state assessment model for processing, obtaining analysis and assessment results and TCM characteristic state labels. Based on the analysis and assessment results and TCM characteristic state labels, a pericardial data analysis report is generated to achieve intelligent health management.

[0026] It should be noted that the data receiving module 110 is connected to the feature processing module 120, the feature processing module 120 is connected to the data acquisition module 130, the data acquisition module 130 is connected to the data fusion module 140, and the data fusion module 140 is connected to the evaluation and report generation module 150. This pericardial health data analysis and processing system is used for intelligent analysis of medical and health data. By intelligently analyzing the collected data, it assists in early medical analysis and can be used for TCM-assisted assessment, abnormal state monitoring, and risk warning. High-frequency sampling is performed through a specific time window to extract quantified pulse characteristics (pulse position, pulse strength, pulse shape, pulse number). Combined with dynamic baselines and multi-dimensional data fusion, the system uses a multimodal physiological state assessment model to output analysis and assessment results along with TCM characteristic state labels. This quantifies the pericardial data and uses the quantified data to assist in diagnosis, providing objective data references for time-based acupuncture and medication.

[0027] The technical solutions provided in the embodiments of this application will be further described below with reference to the accompanying drawings.

[0028] Reference Figure 1 , Figure 1It is a schematic diagram of the processing structure of the pericardial health data analysis and processing system provided by an embodiment of this application. The pericardial health data analysis and processing system is communicatively connected to a wearable device. The wearable device is used to collect the user's first physiological data according to a preset first rule and the user's second physiological data according to a second rule. Among them, the first rule is to sample at a first frequency during a first time period, and the second rule is to sample at a second frequency during a second time period. The first time period corresponds to a physiologically active time period, and the second time period is the time period other than the first time period on the same day. The first frequency is greater than the second frequency.

[0029] Among them, the wearable device is a smart wearable device with a multi-channel PPG sensor. The user wears this smart wearable device. There is an RTC clock built into the wearable device, and a specific enhanced monitoring window is set according to this clock, corresponding to the data collection in the first time period. The first time period corresponds to a physiologically active time period, and the physiologically active time period is the Chinese traditional medicine Xushi. Chinese traditional medicine Xushi is from 19:00 to 21:00 every day. By corresponding the characteristics to Xushi in Chinese traditional medicine and quantifying it through the first time period, it is convenient for subsequent calculations. This time period is based on Beijing time as the reference time. When the user is in different regions, the conversion of the local region is carried out according to this reference time to ensure the consistency in time, so that the collected data can reflect the characteristics of pericardial data. According to traditional Chinese medicine theory, Xushi is the time when the qi and blood of the Hand-Jueyin Pericardium Meridian flow most vigorously, and the functional state of the pericardium is most obvious in the pulse at this time. In order to collect the data during this time period, an RTC clock is built into the wearable device and set in the first time period. The wearable device samples the first physiological data at the first frequency, and the first frequency can be once per minute. The second time period is the remaining time period of the whole day except for Xushi in Chinese traditional medicine. During the second time period, the wearable device samples the second physiological data at the second frequency, and the second frequency is less than the first frequency, which can be once every 5 minutes or once every  10 minutes.

[0030] Different from the data collection of wearable devices in the prior art, the first physiological data of the user in this time period is sampled at a high frequency of the first frequency during the first time period, ensuring that sufficient fine waveform data is obtained (this waveform data can reflect the flow of qi and blood in the Pericardium Meridian in traditional Chinese medicine theory for subsequent calculations), and the second physiological data of the user in the remaining time period is sampled at a low frequency of the second frequency during the second time period, so as to achieve comprehensive data sampling. Through time-division variable-frequency sampling, accurate data is provided for subsequent calculations, increasing the accuracy of data analysis.

[0031] In one embodiment, a module in a pericardial health data analysis and processing system is executed by a processor in an electronic device or a readable storage medium. The pericardial health data analysis and processing system includes a data receiving module 110, a feature processing module 120, a data acquisition module 130, a data fusion module 140, and an evaluation and report generation module 150.

[0032] The data receiving module 110 is used to receive first physiological data and second physiological data collected by the wearable device. The first physiological data includes the raw PPG waveform signal.

[0033] In one embodiment, the first and second physiological data are data collected from the same user at different time periods. Both include raw PPG waveform signals, heart rate (HR), heart rate variability (HRV), and blood oxygen saturation (SpO2), among other indicators. The second physiological data also includes cardiopulmonary coupling (CPC), skin conductance response, and other data that can reflect sleep and emotional state. Based on the collected heart rate, step count, heart rate variability, cardiopulmonary coupling, and their correlation with time, a multimodal data processing method using a transformer model is employed to obtain the sleep state. Based on heart rate fluctuations and skin conductance response, the skin conductance response reflects the level of emotional tension, which is analyzed using a transformer model to obtain the emotional state. Receiving the first and second physiological data provides the data foundation for subsequent calculations. It should be noted that the transformer model is an existing model, applied to the processing of physiological data only with simple parameter adjustments; further details are omitted here.

[0034] The feature processing module 120 is used to perform feature processing on the raw PPG waveform signal to obtain pericardial pulse characteristics.

[0035] In one embodiment, before performing feature processing on the raw PPG waveform signal, data preprocessing is first performed on the raw PPG waveform signal. Data preprocessing includes baseline shifting and high-frequency noise reduction. Baseline shifting refers to removing respiratory influences, and high-frequency noise reduction removes electromyographic interference. These data processing methods are common waveform processing techniques and will not be elaborated upon here. The following describes the processing of the preprocessed raw PPG waveform signal to obtain pericardial pulse characteristics.

[0036] Specifically, the feature processing module 120 includes a first feature analysis submodule, a second feature analysis submodule, a third feature analysis submodule, and a feature fusion submodule. The first feature analysis submodule is used to analyze the pulse position and pulse potential of the PPG raw waveform signal to obtain pulse position and pulse potential features.

[0037] In one embodiment, in Traditional Chinese Medicine (TCM) pulse diagnosis: pulse position (superficial / deep) typically refers to the depth of the pulse, which in the PPG raw waveform signal is mainly reflected in the amplitude and intensity of the signal. The stronger the signal, the easier it is to be captured by the body surface sensor, analogous to a "superficial pulse"; a weak signal requires deep pressure, analogous to a "deep pulse". Pulse strength (weak / strong) typically refers to the strength and duration of the pulse, which in the PPG raw waveform signal is mainly reflected in energy, represented by the area under the curve, i.e., the total volume and duration of blood filling during one heartbeat. By analyzing the PPG raw waveform signal and mapping amplitude and intensity and signal energy to TCM characteristics, data analysis reflecting TCM theory can be achieved, providing a foundation for subsequent health management.

[0038] The pulse position and pulse potential analysis is performed as follows: First, the area under the PPG raw waveform signal within one cardiac cycle is calculated by integration as the pulse energy value. Then, within this cycle, the absolute amplitude of the main peak of the PPG raw waveform signal is calculated to obtain the amplitude value. The average pulse energy value and the average amplitude value of the PPG raw waveform signal in the second physiological data are calculated. The ratio of the above pulse energy value to the average of the user's current pulse energy value is calculated to obtain the energy deviation ratio, which is used to measure the degree of energy deviation. The ratio of the above amplitude value to the average amplitude value of the user on that day is calculated to obtain the amplitude deviation ratio, which is used to measure the degree of amplitude deviation.

[0039] Then, the energy deviation ratio and amplitude deviation ratio are compared with preset deviation thresholds, which have high and low thresholds. For example, the high threshold can be 1.2 and the low threshold can be 0.8. If both the energy deviation ratio and amplitude deviation ratio are greater than the high threshold, the output feature is a strong pulse and a floating pulse position, which is used as the pulse position and pulse position feature. It should be noted that different pulse positions and pulse positions are corresponding to different magnitudes of the energy deviation ratio and amplitude deviation ratio with the high and low thresholds. A constructed mapping table can be consulted. This mapping table reflects the TCM description features corresponding to different calculation results, so as to facilitate subsequent calculations. According to the comparison results, it can reflect the different pulse positions and pulse positions presented by the currently monitored PPG raw waveform signal. The mapping table was constructed by experts based on experience and analysis of a large amount of data. This mapping table records the correspondence between the above numerical comparison magnitudes and TCM feature descriptions, so as to obtain pulse position and pulse position features, and quantify the TCM description into calculable data for data analysis. For example, when both the energy deviation ratio and amplitude deviation ratio are below a low threshold, the output characteristic is a weak pulse and a deep pulse position, which is used as the pulse position and pulse strength characteristics, etc. Other cases are described in detail here. The above-mentioned objective data is mapped with traditional Chinese medicine theory, and intelligent analysis of pericardial data is achieved through quantification, providing a foundation for subsequent calculations.

[0040] The second feature analysis submodule is used to perform pulse shape analysis on the original PPG waveform signal to obtain pulse shape features.

[0041] In one embodiment, the morphological geometric characteristics of the pulse wave are analyzed (mapping the pulse patterns of "stringy", "hesitant", and "intermittent" in traditional Chinese medicine, which directly reflect the elasticity of the blood vessel wall, the smoothness of blood flow and rhythm).

[0042] In one embodiment, the second feature analysis submodule includes a stiffness calculation submodule, a rhythm statistics submodule, a smoothness calculation submodule, and a calculation fusion submodule. The stiffness calculation submodule is used to calculate the waveform stiffness index based on the rising slope of the PPG raw waveform signal and the ratio of height to dicrotic wave time difference, and then quantify it to obtain the vascular wall tension.

[0043] In some possible embodiments of this application, the rising branch of the PPG raw waveform signal represents the segment from the trough to the peak. Two points are selected within this segment; exemplarily, the 10% peak and the 90% peak are selected. The slope of the line connecting these two points is calculated; the larger the slope, the steeper the rise. This slope directly reflects the speed at which blood rushes out of the heart and impacts the blood vessel wall. On the falling branch after the main peak, the position of the dicrotic peak is located using the second derivative method. The time difference between the main peak and the dicrotic peak is calculated, and the ratio of the user's height to the time difference is calculated to obtain the stiffness index, expressed as SI = h / ΔT, where h is height, ΔT is the time difference, and SI is the stiffness index. After weighted summation of the slope and the stiffness index, a normalized mapping is performed, mapping to a value between 1 and 100 to obtain the blood vessel wall tension. A larger value indicates poorer blood vessel wall elasticity, and the mapping corresponds significantly to the 'string pulse' characteristic in Traditional Chinese Medicine, facilitating subsequent determination of pulse shape characteristics. It should be noted that the normalization mapping is used to obtain the maximum and minimum values ​​of the fused PPG original waveform signal through a maximum-minimum value mapping.

[0044] The rhythm statistics submodule is used to perform regularity statistics on the waveform period of the PPG raw waveform signal to obtain irregular rhythms, and to quantify the irregular rhythms to obtain the pulse degree.

[0045] In some possible embodiments of this application, peak detection is performed on the raw PPG waveform signal to identify the positions of all main peaks in each heartbeat cycle, and the time point sequence corresponding to each peak is recorded. Then, the time difference between two adjacent peaks is calculated to obtain a time difference sequence. The median in the time difference sequence is calculated as the baseline heartbeat cycle at the current moment. The median is used instead of the average value to prevent individual outliers from skewing the baseline. Abnormal rhythms are captured based on this baseline heartbeat cycle. Specifically, each time difference in the time difference sequence is traversed and compared with a preset rhythm threshold. If the time difference is greater than the high threshold, it is determined as a pause in heartbeat, corresponding to "intermittent pulse" in traditional Chinese medicine. If the time difference is less than the low threshold, and the next time difference in the subsequent time difference sequence is significantly prolonged, it is determined as a premature beat, corresponding to "knotted pulse" or "abrupt pulse" in traditional Chinese medicine. The total number and density of abnormal times within the 7 PM to 9 PM are counted, and the abnormal proportion is used as the degree of pulse pause to obtain pulse shape characteristics later. The high threshold of the rhythm threshold is obtained by multiplying the reference heart rate cycle by the high parameter, and the low threshold of the rhythm threshold is obtained by multiplying the reference heart rate cycle by the low parameter. For example, the high parameter can be 1.8 and the low parameter can be 0.6, which are set according to empirical values.

[0046] The smoothness calculation submodule is used to calculate the smoothness of the falling branch of the PPG raw waveform signal and quantize it to obtain the blood flow smoothness.

[0047] In some possible embodiments of this application, the smoothness of the waveform edges is mapped (corresponding to the difference between "hesitant pulse" and "slippery pulse" in Traditional Chinese Medicine). A waveform segment from the peak of the main wave to the next trough is extracted from the original PPG waveform signal, i.e., the descending branch data. A Gaussian function or a cubic spline interpolation function is used to interpolate the descending branch data, constructing a smooth curve as the smoothed descending branch data. The root mean square error between the descending branch data and the smoothed descending branch data is calculated; this error represents minute jitters on the waveform. A larger error value indicates a coarser waveform; a smaller error value indicates a smoother waveform. This error value is normalized and mapped to a value between 0 and 100, indicating that a larger value indicates smoother flow, thus obtaining the blood flow smoothness, which is used to subsequently obtain pulse shape characteristics. It should be noted that the normalization mapping specifically obtains the maximum and minimum values ​​of the root mean square error calculated from the original PPG waveform signal, and then calculates the mapping using the maximum-minimum value.

[0048] The computational fusion submodule is used to perform vector fusion processing on blood vessel wall tension, pulse pause degree, and blood flow smoothness to obtain pulse shape features.

[0049] In some possible embodiments of this application, the vascular wall tension, pulse pause degree and blood flow smoothness are represented by an array to generate a three-dimensional feature vector, thereby obtaining pulse shape features, so that the pericardial pulse features can be obtained by subsequent fusion based on the pulse shape features.

[0050] The third feature analysis submodule is used to perform pulse number analysis on the raw PPG waveform signal to obtain pulse number features.

[0051] In one embodiment, pulse count analysis specifically involves: first, peak detection is performed on the raw PPG waveform signal to identify the positions of all main peaks in each heartbeat cycle, and the time point sequence corresponding to each peak is recorded. Then, the time difference between two adjacent peaks is calculated to obtain a time difference sequence. The mean of all values ​​in the time difference sequence is calculated, representing the heart rate per minute. The calculated mean is compared with a preset pulse count threshold, corresponding to a high threshold and a low threshold. If the mean is greater than the high threshold, it is marked as 1 (corresponding to predominantly hot); if the mean is less than the low threshold, it is marked as 0 (corresponding to predominantly cold); otherwise, it is considered normal. For example, the high threshold can be 90 beats / min, and the low threshold can be 60 beats / min, set based on the heart rate of most people. The standard deviation or variance can also be calculated using the time difference sequence to reflect the stability of the pulse. Through the above pulse count analysis, pulse count characteristics are obtained, enabling the analysis of collected objective data and mapping the analysis results to traditional Chinese medicine theory, achieving intelligent analysis. The correspondence between the above comparison results can be obtained by looking up a mapping table.

[0052] The feature fusion submodule is used to perform vector fusion processing on pulse position and pulse momentum features, pulse shape features and pulse number features to obtain pericardial pulse features.

[0053] In one embodiment, pulse position and pulse strength characteristics, pulse shape characteristics, and pulse number characteristics are fused using feature concatenation processing to obtain pericardial pulse characteristics. These pericardial pulse characteristics can be represented by an array, facilitating the extraction of corresponding features. These pericardial pulse characteristics reflect the combination of PPG raw waveform signal analysis and traditional Chinese medicine theory, providing a foundation for subsequent risk assessment analysis.

[0054] The data acquisition module 130 is used to acquire dynamic baseline data, which is obtained by processing the historical physiological data of the user collected by the wearable device through a preset dynamic baseline analysis model.

[0055] In one embodiment, the user's historical physiological data is physiological data collected over a past period via a wearable device. This data, along with the first and second physiological data, belongs to the same user, and the past period can range from two weeks to one month. The dynamic baseline analysis model is a time-decay weighted moving average algorithm. First, the historical physiological data is preprocessed; the preprocessing process is not detailed here. Feature extraction is performed on the historical data to obtain the amplitude and energy values ​​for each day. The amplitude values ​​are weighted, with weights increasing as time progresses, and the sum of the weights is 1. The weighted average is then calculated to obtain the dynamic amplitude baseline. Similarly, the energy values ​​are processed to obtain the dynamic energy baseline. Simultaneously, the user's historical physiological data heart rate is recorded. The dynamic amplitude baseline, dynamic energy baseline, and heart rate are stored in an array format to obtain dynamic baseline data. The dynamic baseline data is obtained through a preset data reading function, which can be either `read()` or `open()`, to provide a reference benchmark for subsequent calculations.

[0056] It should be noted that the dynamic baseline data also includes heart rate, cardiopulmonary coupling, skin conductance, etc., which are similar to the multimodal data processing step S100 to obtain emotional state and vital sign data, in preparation for subsequent calculations.

[0057] It should also be noted that this dynamic baseline data is generated over a period of time prior to the current time and is continuously updated over time. By obtaining dynamic baseline data, it is possible to reflect the user's recent status and provide a more reliable reference.

[0058] The data fusion module 140 is used to fuse pericardial pulse characteristics, second physiological data and dynamic baseline data to obtain a multi-dimensional feature vector.

[0059] In one embodiment, data fusion can enhance the representational ability of data and improve the accuracy of analysis.

[0060] Specifically, the data fusion module 140 includes a weighted fusion submodule, an association calculation submodule, a search submodule, and a vector construction submodule. The weighted fusion submodule is used to perform weighted fusion of pericardial pulse features and second physiological data to obtain a first-dimensional feature vector, wherein the weight of the pericardial pulse features is greater than the weight of the second physiological data.

[0061] In one embodiment, since the pericardial pulse characteristics are obtained from data collected in the first time period (mapped to the period when the Qi and blood of the pericardial meridian are at their peak), they can better reflect the function of the human pericardium. Setting the weight of the pericardial pulse characteristics greater than the weight of the second physiological data not only allows for combination with other dimensions of data to enhance data representation capabilities but also better reflects data characteristics. The sum of the weights of the pericardial pulse characteristics and the second physiological data is 1. For example, the weight of the pericardial pulse characteristics is 0.7, and the weight of the second physiological data is 0.3. Multiplying the pericardial pulse characteristics by their corresponding weights, multiplying the second physiological data by their corresponding weights, and then adding them together yields the first-dimensional feature vector, which is used to subsequently obtain the multi-dimensional feature vector.

[0062] The correlation calculation submodule is used to correlate the pericardial pulse characteristics with the vital signs and status data in the dynamic baseline data to obtain the second-dimensional feature vector.

[0063] like Figure 2 As shown, the correlation calculation submodule includes a feature extraction submodule, an anomaly analysis submodule, and a feature concatenation submodule. The feature extraction submodule is used to extract heart rate and emotional state from vital signs data that are in the same period as the pericardial pulse characteristics.

[0064] In some possible embodiments of this application, vital sign data includes the user's heart rate, emotional state, sleep state, etc. The heart rate and emotional state, occurring simultaneously with the pericardial pulse characteristics, are extracted using a data reading method corresponding to the vital sign data storage format, preparing for subsequent calculations. For example, the reading method corresponding to the storage format is, in the case of array storage, reading the corresponding data through array indices.

[0065] The anomaly analysis submodule is used to perform anomaly analysis on heart rate signs and obtain the anomaly analysis results.

[0066] In some possible embodiments of this application, heart rate is represented by pulse rate. The heart rate is compared to a preset heart rate threshold, which represents the upper and lower limits of a normal pulse rate. A heart rate exceeding the upper limit of the heart rate threshold is marked as an abnormal result, as is a heart rate falling below the lower limit. A heart rate falling between the upper and lower limits of the heart rate threshold is considered normal. The abnormality analysis results obtained through this process provide a basis for subsequently eliminating false alarms caused by physiological activities (such as exercise or excitement).

[0067] The feature splicing submodule is used to determine the confidence level of anomaly analysis results using emotional states.

[0068] The feature splicing submodule is used to splice vital signs data, emotional state and pericardial pulse features to obtain a second-dimensional feature vector when the confidence level is greater than a preset confidence threshold.

[0069] In some possible embodiments of this application, in real life, changes in heart rate and vital signs may also be caused by emotional excitement. Based on the above-mentioned emotional state analysis, confidence is determined by fuzzy matching. The rules for fuzzy matching are: if the emotional state is excited and the movement is still, the confidence is moderate; if the emotional state is stable and the movement is still, the confidence is high; if the emotional state is excited and the movement is still, the confidence is low. The above confidence is mapped to a value between 0 and 1. If the confidence value is greater than a preset confidence threshold, it indicates that there is no abnormality. The vital sign data, emotional state, and pericardial pulse characteristics are concatenated to obtain a second-dimensional feature vector. This second-dimensional feature vector is stored in an array, integrating other vital sign data and emotional state, making the feature representation more comprehensive, accurately reflecting the user's state, and improving the accuracy of the assessment.

[0070] The feature splicing submodule is used to splice vital signs data and pericardial pulse features to obtain a second-dimensional feature vector when the confidence value is less than or equal to a preset confidence threshold.

[0071] In other possible embodiments of this application, if the confidence value is less than a preset confidence threshold, it is considered a non-abnormal situation. The vital signs data and pericardial pulse characteristics are then concatenated to obtain a second-dimensional feature vector. This second-dimensional feature vector is stored in an array, integrating other vital signs data and emotional states, making the feature representation more comprehensive, accurately reflecting the user's state, and improving assessment accuracy.

[0072] The search submodule is used to search for indicator features that are related to the pericardium pulse characteristics in a preset meridian association table, and obtain the third-dimensional feature vector. The meridian association table includes the correspondence between the pericardium pulse characteristics and the preset triple energizer meridian indicators and liver meridian indicators.

[0073] In one embodiment, the pericardium is associated with the internal organs of the body, i.e., they have an exterior-interior relationship, and the states of different organs at different times are representative. The meridian association table links the pericardium meridian function with the indicators of the Triple Energizer meridian and the Liver meridian. Triple Energizer meridian indicators are collected during Hai hour (21:00-23:00), and Liver meridian indicators are collected during Chou hour (01:00-03:00). The meridian association table is obtained by analyzing historical data. The Triple Energizer meridian indicators are represented by body temperature and HRV data, while the Liver meridian indicators are represented by sleep status. The processing method of the above historical data is similar to the method of forming the PPG raw waveform signal, and will not be elaborated here. Based on the analysis of this historical data, the meridian association table is formed through manual annotation and organization. For example, the meridian association table is stored in key-value format, where the key is the pericardium meridian characteristic, and the value is the Triple Energizer meridian indicator and the Liver meridian indicator. The key is a wiry pulse, the Triple Energizer meridian indicator is low HRV, and the Liver meridian indicator is less sleep.

[0074] Based on the current pericardial pulse characteristics, they are matched with the pericardial pulse characteristics recorded in the meridian association table. A text similarity algorithm can be used for matching to obtain the corresponding Sanjiao meridian and liver meridian indicators for the corresponding time period. The corresponding features are then concatenated to obtain a third-dimensional feature vector. This third-dimensional feature vector is represented by an array, considering not only the pericardial data but also the state of its associated data. This matching process achieves comprehensive data coverage, allowing for subsequent improvements in accuracy.

[0075] The vector construction submodule is used to construct a multi-dimensional feature vector from the first-dimensional feature vector, the second-dimensional feature vector, and the third-dimensional feature vector.

[0076] In one embodiment, the first-dimensional feature vector, the second-dimensional feature vector, and the third-dimensional feature vector are fused using feature concatenation to obtain a multi-dimensional feature vector. This multi-dimensional feature vector can be represented using an array or a dictionary, facilitating the extraction of corresponding features. Data fusion increases the comprehensiveness and expressive power of data processing, providing support for accurate risk analysis.

[0077] The assessment and report generation module 150 is used to input multi-dimensional feature vectors into a preset multimodal physiological state assessment model for processing, obtain analysis and assessment results and TCM characteristic state labels, and generate a pericardial data analysis report based on the analysis and assessment results and TCM characteristic state labels.

[0078] In one embodiment, the preset multimodal physiological state assessment model is a pre-trained model, which can be a multi-task deep learning model or a deep learning model with attention heads. Based on the extracted multi-dimensional feature vectors, the multi-dimensional feature vectors are input into the deep learning model with attention heads, and the analysis and assessment results and TCM characteristic state labels are output. The analysis and assessment results are used to characterize the degree of health management risk indicated in the analysis of the multi-dimensional feature vectors, and the TCM characteristic state labels are mapped to TCM-related expressions.

[0079] like Figure 3 As shown, the multimodal physiological state assessment model includes a temporal feature encoder, a semantic feature embedding layer, and a fusion decision layer. The temporal feature encoder is a CNN+Bi-LSTM structure used to extract the first and second dimension feature vectors from the multi-dimensional feature vectors, capturing subtle morphological changes in the pulse. The semantic feature embedding layer is an embedding layer+Self-Attention structure used to extract the third dimension feature vector from the multi-dimensional feature vectors, constructing correlations. The fusion decision layer concatenates the outputs of the temporal feature encoder and the semantic feature embedding layer, and outputs the analysis and assessment results (represented as scores between 0 and 10) and TCM characteristic state labels through a multilayer perceptron.

[0080] Before processing, the multimodal physiological state assessment model is trained. During training, complete PPG data for the Xu hour (7-9 PM) and physiological data for the remaining time period of the day for N users are first acquired as training samples. This data can be saved historical user data, and N can be 2000 or more. For each sample, at least two senior TCM physicians independently diagnose and determine the pericardium meridian function status (e.g., normal, qi deficiency, blood stasis, phlegm obstruction) and the corresponding risk level (0-10) based on the analysis and assessment results. When the opinions of the two physicians are consistent, the label is confirmed as the true label for subsequent loss function calculation. Based on the above process, the collected sample data is processed to form a multidimensional feature vector that can be input into the model (similar to the above process, not elaborated here). The dataset is divided into training and test sets at an 8:2 ratio (other ratios can also be used). A multimodal physiological state assessment model is invoked, and forward propagation is performed. The loss function is calculated using a multi-task loss function, and then backpropagation is performed using this loss function value to adjust the model's parameters. The weight parameters of the CNN and Bi-LSTM layers, as well as the Embedding layer + Self-Attention module, are continuously updated through backpropagation until the loss function converges. This allows the model to learn the nonlinear mapping relationship between waveform features and TCM syndromes. The multi-task loss function can be expressed as Loss = a*L1 + b*L2, where L1 is the classification loss function (using cross-entropy loss) for predicting TCM feature state labels, L2 is the regression loss (using mean squared error) for predicting the risk level of the assessment results, and a and b are balance coefficients. Supervised training can also be performed with manual intervention. The specific training process is similar to that of existing supervised learning models, involving iterative parameter updates, which will not be elaborated here.

[0081] In one embodiment, the evaluation and report generation module includes a label mapping submodule and a report generation submodule. The label mapping submodule is used to map TCM characteristic status labels to preset warning prompts when the analysis and evaluation results exceed the preset risk threshold, so as to obtain status prompt information. In one embodiment, the risk threshold is a risk value set based on historical data statistics, which can be 6 points. If the analysis and evaluation result exceeds the preset risk threshold, it indicates a corresponding risk. The TCM characteristic status label is then mapped to a preset warning prompt to obtain status prompt information. Specifically, a characteristic status label-risk mapping table is obtained. This table includes the correspondence between characteristic status label descriptions and warning prompt information, and is constructed by professionals based on cases. A text similarity algorithm is used to match the TCM characteristic status labels with the characteristic status label descriptions in the characteristic status label-risk mapping table. A similarity result greater than 60% (or other values) indicates a successful match. The corresponding warning prompt information is then used as the status prompt information for subsequent generation of a pericardial data analysis report. For example, if the TCM characteristic status label is "Qi stagnation and blood stasis," the analysis and evaluation result is 7 points, and the found status prompt information is "Excessive heart fire, accompanied by a risk of sleep disorders, easily leading to irritability and palpitations."

[0082] The report generation submodule is used to generate a pericardial data analysis report by combining TCM characteristic status labels and status prompts.

[0083] In one embodiment, TCM characteristic status labels and status prompts are written into a document, which is a pericardial data analysis report. The document can be a Word document or can be represented by illustrations, such as highlighting high-risk content corresponding to status prompts in red or displaying them using a Tai Chi diagram. The pericardial data analysis report provides a visual representation for easy viewing.

[0084] In one embodiment, if the analysis and evaluation result is less than or equal to a preset risk threshold, it indicates that the fluctuation is normal, and a prompt message to maintain a good lifestyle is given. The prompt message to maintain a good lifestyle is written into the pericardial data analysis report for display.

[0085] In one embodiment, the pericardial health data analysis and processing system further includes a suggestion query module and a real-time tracking module. The suggestion query module is used to search for conditioning suggestion information from a preset conditioning plan library after generating a pericardial data analysis report based on the analysis and evaluation results and TCM characteristic status tags. The real-time tracking module is used to push the pericardial data analysis report and conditioning suggestion information to the client and receive feedback on the conditioning effect from the client for real-time tracking.

[0086] Specifically, according to the status prompt information in the pericardium data analysis report, look up the conditioning advice information from the preset conditioning plan library. For example, the advice information can be to recommend acupoint massage, emotional regulation advice, diet advice, and medical guidance corresponding to the first time period (Xu Shi). The preset conditioning plan library is a conditioning plan constructed based on experience according to the analysis of a large number of historical cases and corresponding to different risk warnings, including diet therapy, work and rest, emotion regulation, and acupoint massage, etc.

[0087] Then push the pericardium data analysis report and the conditioning advice to the client. The user can make adjustments according to the conditioning advice and receive the conditioning effect feedback by the client. This conditioning effect is the feeling filled in by the user after the operation. Through the above method, the user's status can be tracked in real time, guidance can be achieved, and intelligent health management can be realized.

[0088] It should also be noted that: when the device provided in the above embodiment realizes its functions, only the above division of each functional module is used for illustration. In actual application, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the system device embodiments provided in the above embodiments belong to the same concept, and the specific implementation process is detailed in the system embodiment, which will not be repeated here.

[0089] This application also discloses an electronic device. Refer to Figure 4 , Figure 4 which is a schematic structural diagram of an electronic device according to an embodiment of this application. The electronic device 500 may include: at least one processor 501, at least one network interface 504, a user interface 503, a memory 505, and at least one communication bus 502.

[0090] Among them, the communication bus 502 is used to realize the connection and communication between these components.

[0091] Among them, the user interface 503 may include a display screen (Display), a camera (Camera). Optionally, the user interface 503 may further include a standard wired interface and a wireless interface.

[0092] Among them, the network interface 504 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0093] The processor 501 may include one or more processing cores. The processor 501 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 505, and by calling data stored in memory 505. Optionally, the processor 501 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array. The processor 501 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and Modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also be implemented as a separate chip without being integrated into the processor 501.

[0094] The memory 505 may include random access memory (RAM) or read-only memory. Optionally, the memory 505 may include a non-transitory computer-readable storage medium. The memory 505 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above embodiments, etc.; the data storage area may store data involved in the above embodiments, etc. Optionally, the memory 505 may also be at least one storage device located remotely from the aforementioned processor 501. (Refer to...) Figure 4 The memory 505, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for a pericardial health data analysis and processing system.

[0095] exist Figure 4In the illustrated electronic device 500, the user interface 503 is mainly used to provide an input interface for the user and to acquire user input data; while the processor 501 can be used to call the application program of an analysis and processing system for pericardial health data stored in the memory 505. When executed by one or more processors 501, the electronic device 500 performs the steps in one or more modules as described in the above embodiments. It should be noted that, for the foregoing embodiments, for the sake of simplicity, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0096] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0097] In the various embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between apparatuses or units may be electrical or other forms.

[0098] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0099] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0100] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0101] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will readily conceive of those skilled in the art upon consideration of the specification and the disclosure of practical truths.

[0102] This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A system for the analysis of pericardial health data, characterized in that, The analysis and processing system is communicatively connected to a wearable device, which is configured to collect first physiological data of a user according to a preset first rule and second physiological data of the user according to a second rule. Among them, the first rule is to sample at a first frequency during a first time period, and the second rule is to sample at a second frequency during a second time period. The first time period corresponds to a physiologically active time period, and the second time period is the time period other than the first time period on the same day. The first frequency is greater than the second frequency. The system includes: A data receiving module, configured to receive the first physiological data and the second physiological data collected by the wearable device. The first physiological data includes a PPG raw waveform signal; A feature processing module, configured to perform feature processing on the PPG raw waveform signal to obtain pericardial pulse characteristics; A data acquisition module, configured to acquire dynamic baseline data, which is obtained by processing historical physiological data of the user collected by the wearable device through a preset dynamic baseline analysis model; A data fusion module, configured to fuse the pericardial pulse characteristics, the second physiological data, and the dynamic baseline data to obtain a multi-dimensional feature vector; An evaluation and report generation module, configured to input the multi-dimensional feature vector into a preset multi-modal physiological state evaluation model for processing to obtain an analysis and evaluation result and a traditional Chinese medicine feature state label, and generate a pericardial data analysis report based on the analysis and evaluation result and the traditional Chinese medicine feature state label.

2. The system of claim 1, wherein, The physiologically active time period is the time of Xu Shi in traditional Chinese medicine. The feature processing module includes a first feature analysis sub-module, a second feature analysis sub-module, a third feature analysis sub-module, and a feature fusion sub-module. The first feature analysis sub-module is configured to perform pulse position and pulse potential analysis on the PPG raw waveform signal to obtain pulse position and pulse potential characteristics; The second feature analysis sub-module is configured to perform pulse shape analysis on the PPG raw waveform signal to obtain pulse shape characteristics; The third feature analysis sub-module is configured to perform pulse rate analysis on the PPG raw waveform signal to obtain pulse rate characteristics; The feature fusion sub-module is configured to perform vector fusion processing on the pulse position and pulse potential characteristics, the pulse shape characteristics, and the pulse rate characteristics to obtain the pericardial pulse characteristics.

3. The system of claim 2, wherein, The second feature analysis sub-module includes a stiffness calculation sub-module, a rhythm statistics sub-module, a smoothness calculation sub-module, and a calculation fusion sub-module. The stiffness calculation sub-module is configured to calculate and quantify the waveform stiffness index based on the slope of the rising branch of the waveform of the PPG raw waveform signal and the ratio of height to dicrotic wave time difference, and obtain the vascular wall tension; The rhythm statistics sub-module is configured to perform regularity statistics on the waveform period of the PPG raw waveform signal to obtain irregular rhythms, and quantify the irregular rhythms to obtain the degree of pulse interruption; The smoothness calculation sub-module is configured to calculate and quantify the smoothness of the falling branch of the waveform of the PPG raw waveform signal to obtain blood flow fluency; The computational fusion submodule is used to perform vector fusion processing on the vessel wall tension, the pulse pause degree, and the blood flow smoothness to obtain the pulse shape feature.

4. The system of claim 1, wherein, The data fusion module includes a weighted fusion submodule, an association calculation submodule, a search submodule, and a vector construction submodule. The weighted fusion submodule is used to perform weighted fusion of the pericardial pulse features and the second physiological data to obtain a first-dimensional feature vector, wherein the weight of the pericardial pulse features is greater than the weight of the second physiological data. The association calculation submodule is used to associate the pericardial pulse characteristics with the vital sign data and status data in the dynamic baseline data to obtain a second-dimensional feature vector. The search submodule is used to search for index features that are associated with the pericardial pulse features in a preset meridian association table, and obtain a third-dimensional feature vector. The meridian association table includes the correspondence between the pericardial pulse features and preset triple energizer meridian indicators and liver meridian indicators. The vector construction submodule is used to construct the multi-dimensional feature vector from the first-dimensional feature vector, the second-dimensional feature vector, and the third-dimensional feature vector.

5. The system of claim 4, wherein, The correlation calculation submodule includes a feature extraction submodule, an anomaly analysis submodule, and a feature concatenation submodule. The feature extraction submodule is used to extract heart rate signs and emotional state from the vital signs data that are in the same period as the pericardial pulse characteristics. The anomaly analysis submodule is used to perform anomaly analysis on the heart rate signs and obtain anomaly analysis results; The feature splicing submodule is used to determine the confidence level of the anomaly analysis result using the emotional state. If the confidence level is greater than a preset confidence threshold, the vital signs data, the emotional state, and the pericardial pulse characteristics are spliced ​​together to obtain the second-dimensional feature vector.

6. The system of claim 1, wherein, The assessment and report generation module includes a label mapping submodule and a report generation submodule. The label mapping submodule is used to map the TCM characteristic status label to a preset warning prompt when the analysis and evaluation result is greater than a preset risk threshold, so as to obtain status prompt information. The report generation submodule is used to generate the pericardial data analysis report by combining the TCM characteristic status labels and the status prompt information.

7. The system of claim 1 or 6, wherein, The analysis and processing system also includes a suggestion query module and a real-time tracking module. The suggestion query module is used to search for conditioning suggestion information from a preset conditioning plan library based on the pericardial data analysis report after the pericardial data analysis report is generated based on the analysis and evaluation results and the TCM characteristic status labels. The real-time tracking module is used to push the pericardial data analysis report and the treatment suggestion information to the client, and receive the treatment effect feedback from the client for real-time tracking.

8. An electronic device, comprising: The system includes a processor, a memory, a user interface, a communication bus, and a network interface. The processor, the memory, the user interface, and the network interface are respectively connected to the communication bus. The memory is used to store instructions. The user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the steps in the system as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the steps in the system as described in any one of claims 1-7.

Images