Traditional Chinese medicine health cloud monitoring control platform data analysis method and system

Abstract

The invention discloses a traditional Chinese medicine health cloud monitoring control platform data analysis method and system, and the method comprises the steps: carrying out the evaluation and cross verification of the logic consistency of medical monitoring equipment hardware working conditions, physiological signal data and subjective symptom description information; and compensation correction and weight adjustment are performed on the data based on the identified fault type, so that the problem of traditional Chinese medicine syndrome vector representation deviation caused by inaccurate data acquisition and data logic contradiction in the prior art is effectively solved, and the accuracy of generating standardized traditional Chinese medicine syndrome vector representation is improved.

Description

Technical Field

[0001] This invention relates to the field of health management data analysis technology, and in particular to a data analysis method and system for a traditional Chinese medicine health cloud monitoring and control platform. Background Technology

[0002] In the field of Traditional Chinese Medicine (TCM) health management, utilizing cloud-based monitoring and control platforms for continuous assessment of an individual's health status is an emerging technological direction. These platforms typically integrate individual physiological signal data automatically collected by wearable medical monitoring devices, as well as subjective feelings and symptom descriptions actively entered by the user through an interactive interface. After receiving this multi-source information, the platform performs a series of processing and analyses, ultimately generating a standardized vector representing the user's current TCM syndrome state, providing a basis for subsequent health interventions.

[0003] However, in actual use, wearable monitoring devices that users use for extended periods experience increased charging contact resistance due to oxidation of the metal contacts on their charging ports. This means that the built-in battery never reaches a fully charged state each time the device is charged. The battery operating in a slightly depleted state for extended periods directly impacts the medical sensors inside the device that collect physiological signals. The low power supply voltage of the sensors causes a small but persistent systematic negative offset during data acquisition. This offset is so subtle that it usually doesn't trigger the platform's outlier alarm threshold because it remains within the normal range of physiological fluctuations, yet it deviates from the user's true state.

[0004] At the same time, users' subjective feelings may deviate from the "objective" data presented by the platform. Under the influence of this cognitive bias, when users enter subjective symptom descriptions on the platform application, their trust in the device leads to incorrect entry of subjective symptom information.

[0005] During data transmission, network congestion can cause intermittent packet loss and retransmission of both the continuous raw physiological signal data packets uploaded by medical sensors and the discrete subjective symptom description information data packets submitted by users. The timestamps of the physiological signal data and subjective symptom description information, which should be closely correlated in time, may be misaligned when they arrive at the platform server. This misalignment undermines the basis for judging the development and changes of symptoms, because the platform cannot determine whether a certain physiological change and a certain subjective symptom occurred at the same time or have a direct causal relationship.

[0006] At the platform's data processing level, current data cleaning rules are mainly used to identify and process outliers that are missing data or significantly exceed preset thresholds. These rules are usually unable to effectively identify persistent, minor, systematic data deviations. Because these deviations are within the normal fluctuation range and persist, they are "allowed" by the cleaning rules.

[0007] Ultimately, the platform's medical syndrome classification standards faced a dual dilemma when receiving this "cleaned" but actually deeply flawed data: on the one hand, the systematically downgraded physiological data might mislead the platform, causing it to tend to draw incorrect conclusions; on the other hand, subjective information incorrectly entered by users further reinforced this erroneous judgment. More importantly, the temporal misalignment between physiological data and chief complaint information undermined the platform's accurate basis for judging the changes in symptoms over time. Under the influence of these complex and interacting problems, the platform ultimately generated a non-standardized TCM syndrome vector representation that deviated from the user's true state based on this original information, which contained systematic biases, content conflicts, and temporal misalignments. This vector representation lost its effectiveness as a basis for health monitoring and intervention because it failed to accurately reflect the user's true health status, thus failing to guide effective health management and intervention measures. Summary of the Invention

[0008] The purpose of this invention is to propose a data analysis method for a TCM health cloud monitoring and control platform. This method aims to solve the problem that existing technologies suffer from complex data pollution caused by the gradual hardware degradation due to physical oxidation of the charging interface of medical monitoring equipment, combined with human cognitive bias and network latency, which prevents the TCM health cloud monitoring and control platform from accurately reflecting the user's true health status.

[0009] The present invention is implemented as follows: a data analysis method for a traditional Chinese medicine health cloud monitoring and control platform, the method comprising the following steps: Acquire users' physiological signal data and subjective symptom descriptions; Based on the charging behavior information of the medical monitoring equipment, the hardware condition of the equipment is evaluated to generate equipment operation status evaluation indicators; the logical consistency of the physiological signal data and the subjective symptom description information is verified to generate data content logical evaluation indicators. Cross-validation judgment is performed based on the medical monitoring equipment operation status evaluation index and the data content logic evaluation index to determine the specific manifestation of data content logic contradictions and match the fault type. The physiological signal data is compensated and corrected based on the fault type, and the weight of the subjective symptom description information in subsequent processing is adjusted. Based on the corrected physiological signal data and the adjusted weighted subjective symptom descriptions, a standardized TCM syndrome vector representation is generated.

[0010] Furthermore, the step of evaluating the hardware condition of the medical monitoring device based on its charging behavior information and generating equipment operating status evaluation indicators includes the following steps: Record charging behavior information of medical monitoring devices; Calculate a reference charging efficiency value based on the charging behavior information; The reference charging efficiency value is compared with the actual unit power consumption time, and a hardware reliability score for the medical monitoring equipment is generated based on the comparison results.

[0011] Furthermore, the charging behavior information includes: medical monitoring device identifier, initial battery percentage, final battery percentage, charging completion timestamp, and charging start timestamp.

[0012] Further, based on the charging behavior information, the reference charging efficiency value is calculated as follows: The average unit power consumption time of the medical monitoring device during the most recent N charging cycles is used as the reference charging efficiency value. Here, N is greater than 1, and the unit power consumption time is calculated as: total charging time / (end battery percentage - start battery percentage).

[0013] Furthermore, the step of comparing the reference charging efficiency value with the actual unit power consumption time, and generating a hardware reliability score for the medical monitoring equipment based on the comparison result, includes... Obtain the actual unit power consumption time of the medical monitoring equipment during its most recent charging; When the actual unit power consumption time exceeds the reference charging efficiency value by a preset ratio, it is determined that the reason for the significant decrease in charging efficiency is a specific hardware operating condition abnormality. Calculate the hardware reliability score, score = 100 - (actual unit power consumption time / reference charging efficiency value - 1) K, where K is an adjustment coefficient.

[0014] Furthermore, the logical consistency verification between the acquired physiological signal data and the subjective symptom description information is performed to generate data content logical evaluation indicators, including the following steps: Pre-establish association rules between physiological and chief complaint symptoms for each user; Based on the received subjective symptom description information, query the associated expected physiological data change trends in the association rules; The trend of the physiological data change is compared with the physiological signal data; Based on the comparison results, a data content logical consistency score is generated.

[0015] Furthermore, the specific manifestation of the logical contradiction in the data content is determined and the fault type is matched by cross-validation judgment based on the medical monitoring equipment operation status evaluation index and the data content logic evaluation index; Includes the following steps: Obtain the hardware reliability score of the medical monitoring device and the logical consistency score of the data content; When the hardware reliability score of the medical monitoring device is less than the first preset limit and the logical consistency score of the data content is less than the second preset limit, the specific manifestation of the logical contradiction in the data content is determined. Match the specific manifestations of the logical contradictions in the content to the fault types of the abnormal operating status of the medical monitoring equipment.

[0016] Furthermore, the step of compensating and correcting the physiological signal data based on the fault type and adjusting the weight of the subjective symptom description information in subsequent processing includes the following steps: The compensation value is determined based on the hardware reliability score of the medical monitoring equipment and the preset calculation method; The physiological signal data is compensated and corrected according to the compensation value to obtain corrected physiological signal data. Adjust the weight of the subjective symptom description information in subsequent processing.

[0017] Furthermore, the step of generating standardized TCM syndrome vector representations includes: Features are extracted from the corrected physiological signal data and the weighted subjective symptom description information; The extracted features are mapped to a preset TCM syndrome dimension space; Based on the mapped features, the standardized TCM syndrome vector representation is generated.

[0018] Another objective of this invention is to provide a data analysis system for a Traditional Chinese Medicine health cloud monitoring and control platform, the system comprising: The data acquisition module is used to acquire users' physiological signal data and subjective symptom descriptions. The equipment status assessment module assesses the hardware condition of the medical monitoring equipment based on its charging behavior information and generates equipment operation status assessment indicators. The data logic comparison module performs logical consistency verification between the physiological signal data and the subjective symptom description information and generates data content logic assessment indicators. The cross-validation judgment processing module is used to perform cross-validation judgment based on the medical monitoring equipment operation status evaluation index and the data content logic evaluation index, to determine the specific manifestation of the data content logic contradiction and match the fault type; The data correction and weight adjustment module is used to compensate and correct the physiological signal data according to the fault type, and adjust the weight of the subjective symptom description information in subsequent processing. The syndrome vector generation module is used to generate standardized TCM syndrome vector representations based on corrected physiological signal data and weighted subjective symptom descriptions.

[0019] Beneficial effects of the present invention This invention proposes a data analysis method and system for a Traditional Chinese Medicine (TCM) health cloud monitoring and control platform. By evaluating and cross-validating the logical consistency between the hardware operating conditions of medical monitoring equipment, physiological signal data, and subjective symptom description information, and by compensating, correcting, and adjusting the weights of the data based on the identified fault types, it effectively solves the problems of inaccurate data acquisition and data logic contradictions leading to deviations in the representation of TCM syndrome vectors in existing technologies. It has the advantage of effectively identifying and correcting data inaccuracies caused by abnormal hardware operating conditions of medical monitoring equipment, deviations in the input of user subjective symptom description information, and logical contradictions between physiological signal data and subjective symptom description information, thereby improving the accuracy and reliability of generating standardized TCM syndrome vector representations. Attached Figure Description

[0020] Figure 1 This is a flowchart of a data analysis method for a TCM health cloud monitoring and control platform according to a preferred embodiment of the present invention; Figure 2 This is a structural diagram of a data analysis system for a TCM health cloud monitoring and control platform, according to a preferred embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. For ease of explanation, only the parts related to the embodiments of this invention are shown. It should be understood that the specific embodiments described herein are merely for explaining this invention and are not intended to limit it. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0022] Traditional TCM health cloud monitoring and control platforms suffer from insufficient accuracy and reliability in processing multi-source health data. Specifically, when medical monitoring devices experience gradual hardware degradation, such as physical oxidation of the charging interface, leading to abnormal power supply to internal sensors and resulting in minute but persistent systemic physiological signal data shifts, the platform struggles to effectively identify and correct these shifts. Simultaneously, users, relying on this skewed data, develop cognitive biases and record subjective symptom information that contradicts their actual feelings, further exacerbating data distortion. Furthermore, independent network transmission issues cause timestamp misalignments between physiological signals and subjective information, preventing the platform from accurately establishing temporal correlations between data. This complex data contamination, resulting from hardware degradation, user cognitive biases, and network latency, renders the generated TCM syndrome vector representation inaccurate in reflecting the user's true health status.

[0023] In this regard, refer to Figure 1 A preferred embodiment of the present invention provides a flowchart of a data analysis method for a traditional Chinese medicine health cloud monitoring and control platform; the method includes the following steps: S1, acquire the user's physiological signal data and subjective symptom description information; S2, based on the charging behavior information of medical monitoring equipment, evaluate the hardware condition of the equipment and generate equipment operation status evaluation indicators; S3, perform logical consistency verification on the physiological signal data and subjective symptom description information, and generate data content logical evaluation index; S4, cross-evaluation based on medical monitoring equipment operation status assessment indicators and data content logical assessment indicators. S5, based on the fault type, compensates and corrects the physiological signal data, and adjusts the weight of subjective symptom description information in subsequent processing; S6 generates a standardized TCM syndrome vector representation based on the corrected physiological signal data and the adjusted weighted subjective symptom description information.

[0024] Physiological signal data refers to quantitative information reflecting the objective physiological state of the human body, automatically collected by medical monitoring equipment, such as body temperature, pulse rate, heart rate, blood pressure, blood oxygen saturation, and skin conductance. This data can be collected using various biosensors, wearable devices, or medical-grade monitors, with the aim of obtaining real-time, objective data on the user's physical condition. Subjective symptom description information refers to descriptions of symptoms or physical conditions actively entered by the user through an interactive interface, based on their own feelings and experiences, such as dry mouth, sore throat, irritability, sleep quality, and dietary preferences. This data can be obtained through text input, preset option selection, or speech recognition-to-text conversion, with the aim of supplementing the user's subjective feelings and symptoms that cannot be directly reflected by physiological signal data.

[0025] Charging behavior information refers to the relevant data records generated by the medical monitoring device during the charging process, including the medical monitoring device identifier, initial battery percentage, final battery percentage, charging completion timestamp, and charging start timestamp. This data can be recorded and uploaded by the device's internal power management module or charging management chip, and its purpose is to provide raw data for assessing the health status of the device's battery and charging system. Device hardware condition assessment refers to the process of analyzing and judging the hardware operating status of the medical monitoring device based on the charging behavior information. This may include calculating indicators such as charging efficiency, battery health, and changes in charging interface contact resistance, with the aim of identifying whether the device has experienced performance degradation or abnormal data acquisition due to hardware aging, wear, or malfunction. Device operating status assessment indicators are numerical values ​​or levels generated after the device hardware condition assessment, quantifying the current hardware health status and data reliability of the medical monitoring device. These can be a reliability score, a status level, or a fault code, and their purpose is to provide a reliable basis at the device level for subsequent data processing.

[0026] Logical consistency verification refers to the process of comparing and verifying whether there are contradictions or inconsistencies between physiological signal data and subjective symptom descriptions. This can be done using pre-defined association rules between physiological data and subjective symptoms, expert knowledge bases, or machine learning models. The data content logical evaluation index is a numerical value or level generated after logical consistency verification, quantifying the degree of logical agreement between physiological signal data and subjective symptom descriptions. It can be a consistency score, a contradiction level, or a conflict type identifier.

[0027] Cross-validation refers to the process of comprehensively analyzing and comparing equipment operating status evaluation indicators and data content logical evaluation indicators. This can be done using decision trees, rule engines, or multi-factor fusion models. Fault type refers to the result of classifying and matching the specific manifestations of logical contradictions in the data content after cross-validation. This can include data offsets caused by equipment hardware malfunctions, subjective information distortion caused by user cognitive biases, time misalignments caused by data transmission delays, or combinations thereof.

[0028] Compensation correction refers to the numerical adjustment or correction of physiological signal data based on the identified fault type to eliminate data distortion caused by equipment hardware malfunctions or systematic biases. This can employ methods such as additive compensation, multiplicative correction, regression correction based on historical data, or model prediction correction. Weight adjustment refers to the dynamic adjustment of the influence of subjective symptom description information on subsequent data processing and syndrome generation based on the identified fault type. This can include reducing its weight, increasing its weight, or completely ignoring its influence.

[0029] The standardized TCM syndrome vector representation refers to the process of transforming corrected and adjusted physiological signal data and subjective symptom descriptions into a numerical vector within a predefined TCM syndrome dimensional space through feature extraction and mapping. This vector can be generated using feature engineering, dimensionality reduction algorithms, deep learning models, or expert system rules. Its purpose is to provide a unified, quantifiable, and computer-processable representation of TCM syndrome states for subsequent health management and intervention.

[0030] In a preferred embodiment of the present invention, a data analysis method for a traditional Chinese medicine health cloud monitoring and control platform is specifically implemented as follows: First, the data acquisition module receives users' physiological signal data via a wireless communication interface, such as continuous heart rate and body temperature data streams from a smart bracelet, and subjective symptom descriptions submitted by users via a mobile application, such as users selecting preset symptom tags like dry mouth and fatigue or entering brief text descriptions. The device status assessment module continuously monitors the charging behavior of the medical monitoring device in the background. For example, when the device is connected to a charger, the system records the charging start timestamp and initial battery percentage; when the device is fully charged or disconnected from charging, it records the charging completion timestamp and ending battery percentage. This charging behavior information is used to calculate the device's unit battery consumption time and compare it with historical averages to generate a hardware reliability score for the medical monitoring device. For example, the score ranges from 0 to 100, with a score below 60 potentially indicating a hardware malfunction. The data logic comparison module pre-establishes a set of personalized physiological and subjective symptom association rules for each user; for example, elevated body temperature is often accompanied by sore throat or dry mouth. When new physiological signal data and subjective symptom descriptions are received, this module queries the association rules based on the subjective symptoms, predicts the expected physiological data change trend, and compares it with the actual physiological signal data. The comparison results are used to generate a data content logical consistency score. For example, the score ranges from 0 to 100, and a score below 70 may indicate a logical contradiction in the data content. The cross-validation judgment processing module obtains the generated hardware reliability score and data content logical consistency score. For example, if the hardware reliability score is lower than a preset first limit and the data content logical consistency score is lower than a preset second limit, the system determines that there is a composite data pollution and identifies its specific manifestation, such as the coexistence of device sensor offset and user cognitive bias, and matches the corresponding fault type, such as abnormal sensor power supply causing low data. The data correction and weight adjustment module performs correction according to the matched fault type. For example, if the fault type is abnormal sensor power supply causing low data, the system will calculate a compensation value based on the hardware reliability score and a preset compensation model, and add it to the original physiological signal data to obtain corrected physiological signal data. At the same time, if it is determined that there is cognitive bias in the user's subjective symptom description information, the weight of this subjective symptom information in subsequent processing will be reduced, for example, from the default weight of 1.0 to 0.5. Finally, the syndrome vector generation module receives corrected physiological signal data and weighted subjective symptom descriptions. This module extracts key features from this data, such as average and fluctuating body temperature, pulse frequency and rhythm, and the frequency and severity of subjective symptoms. These features are then mapped to a predefined TCM syndrome dimension space, such as Yin deficiency, Yang deficiency, Qi deficiency, and blood stasis. Based on the mapped features, a standardized TCM syndrome vector representation is generated. Each component of this vector represents the user's tendency or intensity in a specific TCM syndrome dimension, thus providing a quantitative basis for TCM diagnosis and health management.

[0031] Through the above technical solution, this invention effectively addresses the complex data pollution problem in TCM health cloud monitoring and control platforms caused by the combined effects of medical monitoring equipment hardware degradation, user cognitive bias, and network latency. Specifically, by analyzing device charging behavior, the hardware operating condition of the device is assessed, thereby identifying and quantifying potential data deviations caused by the device itself. Simultaneously, by performing logical consistency checks on physiological signal data and subjective symptom descriptions, user cognitive biases or data entry errors are effectively revealed. Furthermore, by cross-validating device operating status evaluation indicators and data content logical evaluation indicators, this application can accurately distinguish and match fault types from different sources, making data pollution identification more precise. Based on the identified fault types, targeted compensation and correction are applied to physiological signal data, and the weight of subjective symptom descriptions in subsequent processing is dynamically adjusted, significantly improving the accuracy and reliability of the original data. Ultimately, the generated TCM syndrome vector representation more realistically and accurately reflects the user's actual health status, providing a reliable decision-making basis for subsequent health management and intervention, and enhancing the overall effectiveness and user trust of the TCM health cloud monitoring platform.

[0032] In some embodiments of the present invention, a method is proposed to evaluate the hardware condition of medical monitoring devices based on their charging behavior information and generate device operating status evaluation indicators. Specifically, this evaluation can be achieved by simply monitoring broad charging behavior information such as charging connection status, charging duration, or number of charging cycles to determine whether the device is in normal working condition. For example, if the device has not been charged for a long time or has too many charging interruptions, it may be marked as abnormal. This can provide a preliminary identification of the device's operating status. However, in its implementation, relying solely on broad charging behavior information for evaluation cannot accurately quantify the health status of the device hardware, making it difficult to effectively distinguish between normal wear and tear and abnormal operating conditions, thus affecting the accuracy of subsequent data correction and weight adjustment.

[0033] To address this, this invention further proposes to evaluate the hardware operating conditions of medical monitoring devices based on their charging behavior information, and generate device operating status evaluation indicators, including the following steps: S21, records charging behavior information of medical monitoring equipment; S22, Calculate a reference charging efficiency value based on the charging behavior information; S23, compare the reference charging efficiency value with the actual unit power consumption time, and generate a hardware reliability score for the medical monitoring device based on the comparison result.

[0034] The reference charging efficiency value refers to the baseline value of the time required to charge a unit of electricity under normal or ideal operating conditions for medical monitoring equipment. It can be determined by statistical analysis of multiple charging data completed in good condition in the past, such as calculating the average or median, to establish a stable reference standard. The actual unit electricity consumption time refers to the actual time spent by the medical monitoring equipment to add one unit of electricity during a specific charging process. It can be calculated by recording the total time of a single charging and the change in the amount of electricity charged. In some preferred embodiments of the present invention, the hardware condition assessment of the medical monitoring device based on its charging behavior information, and the generation of device operating status assessment indicators, are specifically implemented as follows: First, to record the charging behavior information of the medical monitoring device, the platform can be configured with a data acquisition module. This module can receive charging status updates from the medical monitoring device in real time or periodically, and store these updates, such as charging start time, charging end time, and battery percentage before and after charging, in a dedicated database. Next, to calculate a reference charging efficiency value based on the charging behavior information, the system can set a time window, such as the most recent month, and collect data from all complete charging cycles within this window. For each charging cycle, the time consumed per unit of power is calculated, i.e., the total charging time divided by the change in the percentage of charged power. Then, the system can average the time consumed per unit of power for these single charging cycles to obtain an average time consumed per unit of power as the reference charging efficiency value for the device. Finally, to compare the reference charging efficiency value with the actual time consumed per unit of power and generate a hardware reliability score for the medical monitoring device based on the comparison result, the system will immediately calculate the actual time consumed per unit of power for this charging cycle after the device completes a new charging cycle. The system then compares the actual time consumed per unit of electricity with the previously calculated reference charging efficiency value. If the actual time consumed per unit of electricity is significantly higher than the reference value, for example, exceeding a preset percentage threshold, the system can determine that the device's charging efficiency has decreased, potentially indicating a hardware problem. Based on this deviation, the system can apply a preset function or lookup table to generate a hardware reliability score for the medical monitoring device between 0 and 100, where a lower score indicates poorer hardware reliability, and vice versa.

[0035] Through the above technical solution, the embodiments of the present invention can more accurately reflect the health status of medical monitoring equipment hardware by quantifying charging efficiency, thereby effectively distinguishing between normal wear and tear and abnormal operating conditions. This provides a reliable basis for subsequent data correction and weight adjustment, avoiding the inaccuracies caused by relying solely on broad charging behavior information for evaluation, and thus improving the accuracy and reliability of the entire data analysis method.

[0036] In one specific embodiment of the present invention, each time a medical monitoring device is connected to a charging power source, its internal power management module or the accompanying smart charger can automatically record and upload charging behavior information. For example, when a medical monitoring device is identified as "MED-001" and begins charging, the system records its medical monitoring device identifier as "MED-001" and obtains the current initial battery percentage, for example, 20%. When the device completes charging, the system records the final battery percentage, for example, 90%, and also records the charging completion timestamp, for example, 10:30:00 AM on October 26, 2023, and the charging start timestamp, for example, 8:00:00 AM on October 26, 2023. This recorded charging behavior information is then transmitted to a cloud monitoring and control platform. After receiving this data, the platform can calculate the actual unit power consumption time for this charging based on this information. For example, by dividing the difference between the charging completion timestamp and the charging start timestamp (i.e., the total charging time, which is 2 hours and 30 minutes or 150 minutes in this example) by the difference between the ending percentage of battery charge and the starting percentage of battery charge (i.e., the increase in charging capacity, which is 70% in this example), the actual unit battery charge time for this charging can be obtained as approximately 2.14 minutes / percentage. The platform can pre-store the historical charging data of the medical monitoring device "MED-001" and calculate the average unit battery charge time for its most recent N charging cycles as a reference charging efficiency value. For example, if the reference charging efficiency value of "MED-001" is 1.8 minutes / percentage, and the actual unit battery charge time for this charging cycle is 2.14 minutes / percentage, this indicates that the charging efficiency has decreased. The platform can determine whether this decrease is significant based on a preset percentage threshold, thereby identifying whether there is a specific hardware malfunction, such as battery aging or a decline in charging circuit performance. Ultimately, based on this comparison result, the hardware reliability score of the medical monitoring device "MED-001" can be calculated and generated. This score will be used as part of the device's operational status evaluation index for subsequent data correction and weight adjustment.

[0037] In some embodiments of the present invention, a method for evaluating the hardware condition of medical monitoring devices based on charging behavior information is proposed, thereby generating device operating status evaluation indicators. Specifically, this evaluation of the hardware condition based on the charging behavior information of medical monitoring devices can be achieved by recording the charging behavior information of the devices and calculating a reference charging efficiency value based on this information. For example, the unit power consumption time of the most recent complete charge can be used as a reference, which can preliminarily determine the charging efficiency of the devices. However, in its implementation, the hardware condition of the devices cannot be accurately evaluated solely based on the charging behavior information. For example, the reference charging efficiency value cannot be determined, and therefore, it is impossible to accurately determine whether there are any abnormalities in the device hardware.

[0038] In this regard, embodiments of the present invention further propose calculating a reference charging efficiency value based on the charging behavior information: The average unit power consumption time of the medical monitoring device during the most recent N charging cycles is used as the reference charging efficiency value. Here, N is greater than 1, and the unit power consumption time is calculated as: total charging time / (end battery percentage - start battery percentage).

[0039] Among them, the unit power consumption time refers to the time required for the device to charge one percent of the power, and its purpose is to quantify the efficiency of the charging process; the reference charging efficiency value refers to the benchmark efficiency value used to evaluate the hardware operating conditions of medical monitoring equipment.

[0040] In some preferred embodiments of the present invention, the medical monitoring device can continuously record its charging behavior information, including the medical monitoring device identifier for each charge, the initial battery percentage, the final battery percentage, the charging completion timestamp, and the charging start timestamp. To calculate a reference charging efficiency value, the system can set N to 5, meaning that the data from the five most recent complete charging sessions will be used. Specifically, after the device completes a charge, the system calculates the total charging time based on the charging completion timestamp and the charging start timestamp, and calculates the percentage change in battery capacity based on the final battery percentage and the initial battery percentage. For example, if a charge from 20% to 80% takes 60 minutes, then the unit battery time for this charge is 60 minutes / (80% - 20%) = 1 minute / percentage. The system stores this unit battery time data. When the reference charging efficiency value needs to be updated, the system extracts the unit battery time data from the five most recent charges from the historical record. Assume the unit power consumption time for these five charges is 1.0 minute / percentage, 1.05 minute / percentage, 1.0 minute / percentage, 1.1 minute / percentage, and 0.95 minute / percentage, respectively. The system will add these values ​​together and divide by 5 to obtain the average value. For example, (1.0 + 1.05 + 1.0 + 1.1 + 0.95) / 5 = 1.02 minutes / percentage. This calculated 1.02 minutes / percentage is used as the current reference charging efficiency value. In this way, even if there are some small fluctuations during a single charge, such as slight differences in ambient temperature or charging cable connection, these fluctuations will be smoothed out through multiple averaging, resulting in a more stable and accurate reference value for subsequent hardware condition evaluation.

[0041] The above technical solution provides a stable and representative reference charging efficiency value. This calculation method, based on the average of multiple charging data, effectively avoids the impact of randomness or fluctuations in single charging data on the evaluation results, allowing the reference value to more accurately reflect the actual operating conditions of the medical monitoring equipment hardware. Therefore, when this reference value is compared with the actual unit power consumption time, significant decreases in charging efficiency can be identified more reliably, thus accurately determining whether there are any abnormalities in the equipment hardware. This provides an accurate basis for subsequent equipment operation status evaluation and data correction.

[0042] If we rely solely on reference charging efficiency values ​​and ignore real-time data during the actual charging process, we cannot accurately assess the current hardware status of the device, which in turn affects the accuracy of subsequent data correction and analysis.

[0043] To address this, this invention further proposes a method of comparing a reference charging efficiency value with the actual unit power consumption time, and generating a hardware reliability score for the medical monitoring device based on the comparison results, including: S231, Obtain the actual unit power consumption time of the most recent charging of the medical monitoring equipment; S232, when the actual unit power consumption time exceeds the reference charging efficiency value by a preset ratio, the reason for the significant decrease in charging efficiency is determined to be a specific hardware operating condition abnormality. Calculate the hardware reliability score, score = 100 - (actual unit power consumption time / reference charging efficiency value - 1) K, where K is an adjustment coefficient.

[0044] Among them, the actual unit power consumption time refers to the time consumed by the medical monitoring device to charge one unit power percentage during the most recent charging process. It can be calculated by dividing the total charging time by the difference between the final battery power percentage and the initial battery power percentage. The preset ratio is a threshold used to judge whether the difference between the actual unit power consumption time and the reference charging efficiency value reaches a significant level. It can be set based on historical data analysis, expert experience, or device type. Specific hardware operating condition anomalies refer to specific hardware problems that cause a significant decrease in the charging efficiency of the medical monitoring device, such as battery aging, charging interface oxidation, charging circuit failure, etc. It can be categorized according to device model, service life, environmental factors, etc. The hardware reliability score is an indicator that quantitatively evaluates the hardware reliability of the medical monitoring device. It can be a value between 0 and 100, with a higher value indicating more reliable hardware. The adjustment coefficient K is a parameter used to adjust the degree of deviation in the hardware reliability score calculation formula. It can be set according to different device types, application scenarios, or requirements for hardware reliability sensitivity.

[0045] In some preferred embodiments, the present invention is implemented as follows. Assume a medical monitoring device has obtained a reference charging efficiency value by calculating the average time consumed per unit of power during multiple past charging cycles, for example, 10 seconds per unit of power. When the medical monitoring device undergoes its most recent charging, the system monitors and acquires its charging data in real time, calculating the actual time consumed per unit of power for this charging cycle. For example, if the charging cycle from 20% to 80% takes 600 seconds, then the actual time consumed per unit of power is 600 seconds / (80% - 20%) = 600 seconds / 60 = 10 seconds per unit of power. Subsequently, the system compares the actual time consumed per unit of power (10 seconds) with the reference charging efficiency value (10 seconds). If the difference between the actual time consumed per unit of power and the reference charging efficiency value is within a preset ratio, for example, a preset ratio of 10%, i.e., the actual time consumed does not exceed 1.1 times the reference value, then the device hardware is considered to be operating normally. However, if the actual charging time per unit of charge increases due to battery aging or charging interface oxidation, for example, if the calculated actual charging time per unit of charge is 12 seconds, then 12 seconds exceeds the reference charging efficiency value of 10 seconds, and the excess is (12-10) / 10 = 0.2, or 20%. If the preset ratio is set to 15%, then 20% exceeds the preset ratio of 15%, and the system will determine that the significant decrease in charging efficiency is due to a specific hardware malfunction, such as battery performance degradation. Based on this, the system will calculate a hardware reliability score. Assuming the adjustment coefficient K is set to 50, then the hardware reliability score = 100 - (actual charging time per unit of charge / reference charging efficiency value - 1) K = = = = 100 - 10 = 90. This confidence score of 90 can be used as a quantitative assessment of the device's hardware status, indicating a slight performance degradation but still within a usable state. If the actual time per unit of power consumption increases further, for example, to 15 seconds, the score will be lower. = = 100 - 25 = 75, which may indicate a more serious hardware problem. In this way, the platform is able to dynamically assess the hardware health of the medical monitoring device 100 based on real-time charging data.

[0046] Through the above technical solution, the embodiments of the present invention can acquire the actual charging efficiency data of medical monitoring devices in real time and compare it with historical reference values, thereby accurately identifying subtle changes in the device's hardware operating conditions, such as decreased charging efficiency caused by battery aging or charging interface oxidation. This method avoids the limitations of relying solely on historical data while ignoring the device's current true state, and can promptly detect and quantify abnormalities in the device's hardware. By converting the comparison results into quantifiable hardware reliability scores, an evaluation basis is provided for subsequent data correction and analysis, effectively solving the problem of decreased data accuracy caused by device hardware degradation, and ensuring the accuracy of the platform's assessment of the user's health status.

[0047] In further embodiments of the present invention, some of the above-mentioned implementations propose to perform logical consistency verification on physiological signal data and subjective symptom description information, and generate data content logical evaluation indicators. Specifically, the logical consistency verification can be performed by simple rule matching or threshold judgment to initially assess the rationality of the data. This can provide preliminary control over the basic quality of the input data. However, in its implementation, performing only a preliminary logical consistency verification may not be able to deeply identify logical inconsistencies between physiological signal data and subjective symptom description information. These inconsistencies may be caused by a combination of factors. For example, the user's subjective symptom description information may be biased, and the physiological signal data may also be affected by various factors, leading to logical inconsistencies between the two and affecting the accuracy of subsequent data analysis.

[0048] To address this, this invention further proposes a method for logically verifying the consistency between the physiological signal data and the subjective symptom description information, generating logical evaluation indicators for the data content, including: S31, first establish the association rules between physiological and chief complaint symptoms for each user; S32, based on the received subjective symptom description information, query the associated expected physiological data change trends in the association rules; S33, compare the trend of the physiological data change with the physiological signal data; S34. Based on the comparison results, generate a data content logical consistency score.

[0049] Among them, the physiological and chief complaint symptom association rule refers to a pre-established knowledge base or model that reflects the relationship between specific subjective symptoms and expected physiological data changes. It can be implemented using expert experience rule sets based on traditional Chinese medicine theory, statistical models or machine learning models. Among them, the expected physiological data change trend refers to the change pattern or range of physiological data corresponding to the symptom obtained from the physiological and chief complaint symptom association rules based on the user's subjective symptom description information. It can be manifested as the increase, decrease, fluctuation range or specific waveform characteristics of specific physiological indicators. Among them, the data content logical consistency score refers to the numerical value that quantifies the degree of logical consistency between physiological signal data and subjective symptom description information after comparison. It can be a percentage, a value within a range, or a discrete level. Its purpose is to intuitively reflect the data quality and provide a basis for subsequent data correction and weight adjustment.

[0050] In some preferred embodiments, the logical consistency verification of the acquired physiological signal data and subjective symptom description information, and the generation of data content logical evaluation indicators, can be implemented as follows: First, establish pre-defined rules for associating physiological symptoms with subjective complaints for each user. For example, a database can be built containing multiple rules, each defining the correspondence between specific subjective symptoms (such as "dry mouth," "sore throat," and "irritability") and expected physiological data (such as body temperature, pulse rate, and skin conductance). For instance, a rule could define "dry mouth" as typically associated with a slightly elevated body temperature and an increased pulse rate. These rules can be personalized and optimized based on TCM clinical guidelines, expert consensus, and the user's past health data.

[0051] Next, when the system receives subjective symptom descriptions submitted by the user, such as "feeling hot and having a dry mouth," the system queries the established association rules between physiological functions and the user's subjective symptoms. Based on the query results, the system can determine the expected trends in physiological data associated with the symptoms of "fever" and "dry mouth," for example, the expected increase in body temperature and the expected increase in pulse rate.

[0052] The system then compares the actual collected physiological signal data, such as real-time body temperature and pulse data uploaded by the user's medical monitoring device, with the expected physiological data change trends obtained from the query. For example, if the expected body temperature is elevated, but the actual body temperature data shows normal or low, a logical inconsistency is considered to exist. This comparison can employ various algorithms; for instance, it can calculate the deviation between the actual data and the expected trend, or use pattern recognition algorithms to determine the degree of matching.

[0053] Finally, based on the comparison results, the system generates a data content logical consistency score. For example, if the actual body temperature is highly consistent with the expected temperature trend, and the pulse rate also meets expectations, a higher score, such as 95, can be generated; if the actual body temperature deviates significantly from the expected trend, the score will be lowered accordingly, for example, to 60. This score can serve as a reference for subsequent data processing; for example, if the score falls below a certain threshold, data correction or adjustment of the weight of the data source can be triggered.

[0054] In some embodiments described above, this application proposes cross-validation judgment based on medical monitoring device operating status evaluation indicators and data content logic evaluation indicators to determine the specific manifestations of data content logical contradictions and match fault types. Specifically, this judgment can be achieved by analyzing the logical consistency between the device hardware's operating data and physiological signals with subjective symptoms to initially identify potential data anomalies. For example, when the device's charging efficiency significantly decreases or physiological data shows a clear discrepancy with subjective descriptions, the system will mark the data as potentially problematic. This allows for preliminary screening and early warning of data quality. However, in practice, relying solely on a single evaluation indicator is insufficient to accurately determine the specific manifestations of data content logical contradictions, thus hindering effective fault type matching. For instance, when a minor hardware malfunction in the medical monitoring device causes a slight deviation in the data, the data content logical consistency may still be high, leading to misjudgment. Furthermore, the manifestations of data content logical contradictions are diverse, requiring comprehensive analysis combining device operating status and data logical consistency to more accurately determine the fault type.

[0055] To address this, this invention further proposes a method for cross-validating and determining the specific manifestations of logical contradictions in data content and matching fault types based on the medical monitoring device's operational status evaluation indicators and the data content logic evaluation indicators. This includes the following steps: S41, Obtain the hardware reliability score of the medical monitoring device and the logical consistency score of the data content; S42, when the hardware reliability score of the medical monitoring device is less than the first preset limit and the logical consistency score of the data content is less than the second preset limit, the specific manifestation of the logical contradiction in the data content is determined. S43, Match the fault type of the abnormal operating status of the medical monitoring device according to the specific manifestation of the logical contradiction in the content.

[0056] Among them, the hardware reliability score of medical monitoring equipment refers to the quantitative assessment of the reliability of the hardware operation status of medical monitoring equipment, which can be generated by various methods such as device charging behavior, sensor self-test results, or internal diagnostic data; the data content logical consistency score refers to the quantitative assessment of the degree of logical correlation between physiological signal data and subjective symptom description information, which can be generated by methods such as comparison of preset association rules, analysis of machine learning models, or judgment by expert systems; the first preset limit is a threshold used to judge whether the hardware reliability of medical monitoring equipment is in an abnormal state, which can be set based on historical data analysis, equipment performance standards, or clinical experience; the second preset limit... This refers to the threshold used to determine whether the logical consistency of data content is in an abnormal state. It can be set according to data quality requirements, the strength of the correlation between symptoms and physiological data, or expert consensus. The specific manifestation of logical contradiction in data content refers to the specific pattern or type of inconsistency between physiological signal data and subjective symptom description information. It can be a description of the direction of data deviation, duration, or correlation with specific symptoms. The fault type of abnormal operation of medical monitoring equipment refers to the specific hardware or software problems of medical monitoring equipment that cause logical contradiction in data content. It can include sensor failure, power management abnormality, or data transmission error, etc.

[0057] In some preferred embodiments, the present invention is implemented as follows: Assume a user-worn medical monitoring device has a charging port that has developed slight oxidation due to prolonged use, leading to a decrease in battery charging efficiency and consequently affecting the power supply stability of the body temperature sensor. The platform first obtains the hardware reliability score of the medical monitoring device. For example, by analyzing its charging behavior information, it calculates that the actual unit power consumption time is significantly higher than the reference charging efficiency value, thus generating a lower hardware reliability score, for example, 35 points. Simultaneously, the user complains of symptoms of "dry mouth and sore throat," but the body temperature data collected by the device consistently shows 36.5 degrees Celsius, contradicting the user's subjective experience. The platform, through preset physiological and symptom association rules, compares the body temperature data with the expected physiological data change trend of the "heatiness" symptoms, finding a discrepancy, thus generating a lower data content logical consistency score, for example, 40 points. The system presets a first limit of 50 points and a second limit of 45 points. At this point, because the hardware reliability score of the medical monitoring equipment (35 points) is less than the first preset limit of 50 points, and the data content logical consistency score (40 points) is less than the second preset limit of 45 points, the system determines that the cross-validation conditions are met. Furthermore, the system determines that the specific manifestation of the data content logical contradiction is "persistently low physiological data (body temperature) inconsistent with subjective symptoms (heatiness, dry mouth, sore throat)". Finally, based on this specific manifestation of the content logical contradiction, the system matches the fault type of the abnormal operating status of the medical monitoring equipment. For example, the manifestation of "persistently low physiological data (body temperature) inconsistent with subjective symptoms (heatiness, dry mouth, sore throat)" is matched as the fault type of "temperature sensor calibration deviation" or "unstable power supply to the device causing sensor output offset". In this way, the system can accurately identify the underlying cause of the data anomaly, providing clear guidance for subsequent data correction and weight adjustment.

[0058] In some embodiments described above in this application, a method is proposed to compensate and correct physiological signal data based on the identified fault type and adjust the weight of subjective symptom description information in subsequent processing. Specifically, this compensation and correction of physiological signal data based on the identified fault type and adjustment of the weight of subjective symptom description information can be achieved by analyzing fault reports, such as equipment fault codes or user feedback, and then uniformly compensating the physiological signal data according to a preset fault-correction mapping table, and assigning a fixed weight to the subjective symptom description information that is associated with the fault type. For example, if a sensor fault is identified, a fixed offset is uniformly added to all physiological data and the weight of subjective symptom information is reduced. This can initially address equipment faults and user input deviations. However, in its implementation, it merely uses medical monitoring equipment to collect physiological signal data and combines it with the user's subjective symptom description information to generate a standardized TCM syndrome vector representation. Medical monitoring equipment may have inaccurate data due to hardware aging or other issues, and users may also provide inaccurate subjective symptom descriptions due to excessive trust in the equipment data. These factors can cause the generated TCM syndrome vector representation to deviate from the user's true health status.

[0059] To address this, this invention further proposes a method for compensating and correcting the physiological signal data based on the fault type, and adjusting the weight of the subjective symptom description information in subsequent processing, including the following steps: S51, determine the compensation value based on the hardware reliability score of the medical monitoring equipment and the preset calculation method; S52, the physiological signal data is compensated and corrected according to the compensation value to obtain the corrected physiological signal data; S53, Adjust the weight of the subjective symptom description information in subsequent processing.

[0060] In this embodiment of the invention, the standardized TCM syndrome vector representation refers to transforming corrected and adjusted physiological signal data and subjective symptom description information into a numerical vector in a preset TCM syndrome dimensional space through feature extraction and mapping. This vector can be generated using feature engineering, dimensionality reduction algorithms, deep learning models, or expert system rules. Its purpose is to provide a unified, quantitative, and computer-processable representation of TCM syndrome states to facilitate subsequent health management and intervention.

[0061] In some embodiments of the present invention, a standardized TCM syndrome vector representation is generated based on corrected physiological signal data and adjusted subjective symptom description information. Specifically, this standardized TCM syndrome vector representation can be generated by directly applying preset rules or simple models to initially integrate the corrected data and adjusted information. For example, the numerical range of certain physiological indicators can be associated with specific syndromes, or the frequency of subjective symptoms can be linked with the intensity of the syndrome. This can initially achieve a quantitative representation of TCM syndromes. However, in its implementation, it is only a preliminary integration based on these corrected data and adjusted information, lacking the mining of deep features of the data and precise mapping with the TCM syndrome dimensional space. This may result in insufficient accuracy and precision in the generated vector representation, failing to fully reflect the user's complex TCM syndrome state, and making the final health intervention effect less than ideal.

[0062] In some preferred embodiments, the present invention is implemented as follows. Assume a user-worn medical monitoring device 100 has a hardware reliability score of 70 points determined through charging behavior evaluation. This indicates that the device may have some hardware performance degradation, such as a slight systematic deviation in sensor output. The system can preset a calculation method, for example, setting the compensation value as a coefficient proportional to (100 - hardware reliability score) multiplied by a baseline offset. Specifically, if the preset calculation method is: The maximum compensation is 0.5 degrees Celsius. Therefore, for a hardware reliability score of 70, the compensation value can be calculated as follows: Celsius.

[0063] Next, the system compensates and corrects the user's currently collected physiological signal data based on this determined compensation value. For example, if the original body temperature data is 36.5 degrees Celsius, then the corrected physiological signal data can become 36.5 + 0.15 = 36.65 degrees Celsius. In this way, even if there is a slight negative offset in the device, the corrected data can more accurately reflect the user's true body temperature.

[0064] Simultaneously, the system adjusts the weight of subjective symptom descriptions in subsequent processing. Considering the low hardware reliability score (70 points), the system can correspondingly reduce the weight of subjective symptom descriptions. For example, if the normal weight is 1.0, the system can adjust it to 0.8. This means that when generating standardized TCM syndrome vector representations, the system will place greater emphasis on corrected physiological signal data, while appropriately reducing its reliance on subjective symptom descriptions. This effectively avoids the negative impact of users providing inaccurate subjective symptom descriptions due to excessive trust in device data. In this way, even when both device performance degrades and user cognitive biases coexist, the system can generate more accurate TCM syndrome vector representations.

[0065] Through the above technical solution, this application can dynamically determine the compensation value of physiological signal data based on the hardware reliability score of medical monitoring equipment and correct the data, thereby effectively correcting the deviation of physiological signal data caused by hardware aging or failure. Simultaneously, embodiments of this invention can adjust the weight of subjective symptom description information in subsequent processing based on equipment reliability, thereby reducing the negative impact of users providing inaccurate subjective symptom descriptions due to excessive reliance on equipment data. This enables the system to more accurately reflect the user's true health status when generating standardized TCM syndrome vector representations, improving the reliability of syndrome judgment.

[0066] In some embodiments of the present invention, a standardized TCM syndrome vector representation is generated based on corrected physiological signal data and adjusted subjective symptom description information. Specifically, this standardized TCM syndrome vector representation can be generated by directly applying preset rules or simple models to initially integrate the corrected data and adjusted information. For example, the numerical range of certain physiological indicators can be associated with specific syndromes, or the frequency of subjective symptoms can be linked with the intensity of the syndrome. This can initially achieve a quantitative representation of TCM syndromes. However, in its implementation, it is only a preliminary integration based on these corrected data and adjusted information, lacking the mining of deep features of the data and precise mapping with the TCM syndrome dimensional space. This may result in insufficient accuracy and precision in the generated vector representation, failing to fully reflect the user's complex TCM syndrome state, and making the final health intervention effect less than ideal.

[0067] In this regard, embodiments of the present invention further propose steps for generating standardized TCM syndrome vector representations, including: S61, extract features from the corrected physiological signal data and the weighted subjective symptom description information; S62, map the extracted features to a preset TCM syndrome dimension space; S63, Based on the mapped features, generate the standardized TCM syndrome vector representation. In a preferred embodiment of the present invention, the specific process of generating a standardized TCM syndrome vector representation can be implemented as follows.

[0068] First, from the corrected physiological signal data, features such as time-domain and frequency-domain indicators of heart rate variability (HRV), average and fluctuating body temperature, and stability of respiratory rate can be extracted. Simultaneously, from the weighted subjective symptom descriptions, natural language processing techniques can be used to extract semantic features such as the frequency of symptom keywords, symptom severity scores, and correlations between symptoms. For example, for symptoms like "dry mouth" and "sore throat," the frequency of their occurrence in the descriptive text and the severity scores given by users can be extracted. Next, these extracted features are mapped to a predefined TCM syndrome dimension space. This dimension space can be a 12-dimensional vector space defined by a TCM expert system or a machine learning model trained on a large amount of clinical data (e.g., support vector machine or deep neural network), where each dimension corresponds to a major TCM syndrome type, such as "Qi deficiency syndrome," "blood stasis syndrome," and "damp-heat syndrome." The mapping process can employ dimensionality reduction techniques such as Principal Component Analysis (PCA) or Linear Discriminant Analysis (LDA) to project the high-dimensional original feature vectors onto this pre-defined 12-dimensional TCM syndrome space. Alternatively, a pre-trained neural network model can directly output the activation values ​​of the corresponding syndrome dimensions. Finally, based on the mapped features, standardized TCM syndrome vector representations are generated. This can be achieved using a min-max normalization method, scaling the values ​​in each dimension to between 0 and 1, making the values ​​between different syndrome dimensions comparable. For example, if a user's original mapping value is 0.8 in the "Qi Deficiency Syndrome" dimension and 0.2 in the "Damp-Heat Syndrome" dimension, after standardization, they still maintain a relative proportion, but with a unified numerical range. Thus, the final vector, such as [0.7, 0.1, 0.5, ..., 0.2], represents the user's current standardized position in the TCM syndrome dimension space, clearly quantifying their tendency towards each syndrome.

[0069] This invention effectively addresses the complex data contamination problem in Traditional Chinese Medicine (TCM) health cloud monitoring and control platforms caused by the combined effects of medical monitoring equipment hardware degradation, user cognitive biases, and network latency. Specifically, by analyzing device charging behavior, the hardware operating condition is assessed, thereby identifying and quantifying potential data deviations caused by the device itself. Simultaneously, by performing logical consistency checks on physiological signal data and subjective symptom descriptions, user cognitive biases or data entry errors are effectively revealed. Furthermore, by cross-validating device operating status evaluation indicators and data content logical evaluation indicators, this application can accurately distinguish and match fault types from different sources, making data contamination identification more precise. Based on the identified fault types, targeted compensation and correction are applied to physiological signal data, and the weight of subjective symptom descriptions in subsequent processing is dynamically adjusted, significantly improving the accuracy and reliability of the original data. Ultimately, the generated TCM syndrome vector representation more realistically and accurately reflects the user's actual health status, providing a reliable decision-making basis for subsequent health management and intervention, and enhancing the overall effectiveness and user trust of the TCM health cloud monitoring platform.

[0070] Figure 2 This is a structural diagram of a data analysis system for a traditional Chinese medicine health cloud monitoring and control platform according to a preferred embodiment of the present invention. The system includes: The data acquisition module is used to acquire users' physiological signal data and subjective symptom descriptions. The equipment status assessment module assesses the hardware condition of the medical monitoring equipment based on its charging behavior information and generates equipment operation status assessment indicators. The data logic comparison module performs logical consistency verification between the physiological signal data and the subjective symptom description information and generates data content logic assessment indicators. The cross-validation judgment processing module is used to perform cross-validation judgment based on the medical monitoring equipment operation status evaluation index and the data content logic evaluation index, to determine the specific manifestation of the data content logic contradiction and match the fault type; The data correction and weight adjustment module is used to compensate and correct the physiological signal data according to the fault type, and adjust the weight of the subjective symptom description information in subsequent processing. The syndrome vector generation module is used to generate standardized TCM syndrome vector representations based on corrected physiological signal data and weighted subjective symptom descriptions. The data acquisition module, equipment status assessment module, data logic comparison module, cross-validation judgment processing module, data correction and weight adjustment module, and syndrome vector generation module are all specific functional logic processing units within the system. These can be implemented through software programs, hardware circuits, or a combination of both. For example, they can be independent microservices, function libraries, application-specific integrated circuits (ASICs), or programmable gate arrays (FPGAs). The purpose is to decompose complex data analysis methods into manageable and reusable functional units, thereby improving the system's modularity and maintainability.

[0071] The embodiments of the present invention can comprehensively acquire, evaluate, compare, correct, and generate multi-source heterogeneous data, ensuring that even under complex interferences such as device hardware malfunctions, user subjective input deviations, or data transmission delays, a standardized TCM syndrome vector representation that accurately reflects the user's true health status can be generated, significantly improving the accuracy and reliability of data analysis on the TCM health cloud monitoring and control platform.

[0072] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by program instructions and related hardware. The program can be stored in a computer-readable storage medium, such as ROM, RAM, disk, optical disk, etc.

[0073] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A data analysis method for a traditional Chinese medicine health cloud monitoring and control platform, characterized in that, The method includes the following steps: Acquire users' physiological signal data and subjective symptom descriptions; Based on the charging behavior information of medical monitoring equipment, the hardware condition of the equipment is evaluated, and equipment operation status evaluation indicators are generated. The physiological signal data and the subjective symptom description information are logically consistent to generate logical evaluation indicators for data content. Cross-validation judgment is performed based on the medical monitoring equipment operation status evaluation index and the data content logic evaluation index to determine the specific manifestation of data content logic contradictions and match the fault type. The physiological signal data is compensated and corrected based on the fault type, and the weight of the subjective symptom description information in subsequent processing is adjusted. Based on the corrected physiological signal data and the adjusted weighted subjective symptom descriptions, a standardized TCM syndrome vector representation is generated.

2. The data analysis method for a traditional Chinese medicine health cloud monitoring and control platform according to claim 1, characterized in that, The process of evaluating the hardware condition of medical monitoring devices based on their charging behavior information and generating equipment operating status evaluation indicators includes the following steps: Record charging behavior information of medical monitoring devices; Calculate a reference charging efficiency value based on the charging behavior information; The reference charging efficiency value is compared with the actual unit power consumption time, and a hardware reliability score for the medical monitoring equipment is generated based on the comparison results.

3. The data analysis method for a traditional Chinese medicine health cloud monitoring and control platform according to claim 2, characterized in that, The charging behavior information includes: medical monitoring device identifier, initial battery percentage, final battery percentage, charging completion timestamp, and charging start timestamp.

4. The data analysis method for a traditional Chinese medicine health cloud monitoring and control platform according to claim 3, characterized in that, The reference charging efficiency value is calculated based on the charging behavior information: The average time per unit of power consumed during the most recent N charging cycles of the medical monitoring device is used as the reference charging efficiency value. Here, N is greater than 1, and the calculation method for the time per unit of power consumed is: total charging time / (end battery percentage - start battery percentage).

5. The data analysis method for a traditional Chinese medicine health cloud monitoring and control platform according to claim 4, characterized in that, The reference charging efficiency value is compared with the actual unit power consumption time. Based on the comparison result, a hardware reliability score for the medical monitoring equipment is generated. Obtain the actual unit power consumption time of the medical monitoring equipment during its most recent charging; When the actual unit power consumption time exceeds the reference charging efficiency value by a preset ratio, it is determined that the reason for the significant decrease in charging efficiency is a specific hardware operating condition abnormality. Calculate the hardware reliability score, score = 100 - (actual unit power consumption time / reference charging efficiency value - 1)K, where K is an adjustment coefficient.

6. The data analysis method for a traditional Chinese medicine health cloud monitoring and control platform according to claim 1, characterized in that, The logical consistency verification between the acquired physiological signal data and the subjective symptom description information, and the generation of data content logical evaluation indicators, include the following steps: Pre-establish association rules between physiological and chief complaint symptoms for each user; Based on the received subjective symptom description information, query the associated expected physiological data change trends in the association rules; The trend of the physiological data change is compared with the physiological signal data; Based on the comparison results, a data content logical consistency score is generated.

7. The data analysis method for a traditional Chinese medicine health cloud monitoring and control platform according to claim 1, characterized in that, The method involves cross-validating the medical monitoring equipment's operational status assessment indicators and the data content logic assessment indicators to determine the specific manifestations of data content logic contradictions and match fault types. Includes the following steps: Obtain the hardware reliability score of the medical monitoring device and the logical consistency score of the data content; When the hardware reliability score of the medical monitoring device is less than the first preset limit and the logical consistency score of the data content is less than the second preset limit, the specific manifestation of the logical contradiction in the data content is determined. Match the specific manifestations of the logical contradictions in the content to the fault types of the abnormal operating status of the medical monitoring equipment.

8. The data analysis method for a traditional Chinese medicine health cloud monitoring and control platform according to claim 1, characterized in that, Based on the fault type, compensating and correcting the physiological signal data and adjusting the weight of the subjective symptom description information in subsequent processing includes the following steps: The compensation value is determined based on the hardware reliability score of the medical monitoring equipment and the preset calculation method; The physiological signal data is compensated and corrected according to the compensation value to obtain corrected physiological signal data. Adjust the weight of the subjective symptom description information in subsequent processing.

9. The data analysis method for a traditional Chinese medicine health cloud monitoring and control platform according to claim 1, characterized in that, The steps for generating standardized TCM syndrome vector representations include: Features are extracted from the corrected physiological signal data and the weighted subjective symptom description information; The extracted features are mapped to a preset TCM syndrome dimension space; Based on the mapped features, the standardized TCM syndrome vector representation is generated.

10. A data analysis system for a traditional Chinese medicine health cloud monitoring and control platform, characterized in that, The system includes: The data acquisition module is used to acquire users' physiological signal data and subjective symptom descriptions. The equipment status assessment module assesses the hardware condition of the medical monitoring equipment based on its charging behavior information and generates equipment operation status assessment indicators. The data logic comparison module performs logical consistency verification between the physiological signal data and the subjective symptom description information and generates data content logic assessment indicators. The cross-validation judgment processing module is used to perform cross-validation judgment based on the medical monitoring equipment operation status evaluation index and the data content logic evaluation index, to determine the specific manifestation of the data content logic contradiction and match the fault type; The data correction and weight adjustment module is used to compensate and correct the physiological signal data according to the fault type, and adjust the weight of the subjective symptom description information in subsequent processing. The syndrome vector generation module is used to generate standardized TCM syndrome vector representations based on corrected physiological signal data and weighted subjective symptom descriptions.

Images