Traditional Chinese medicine external treatment control method based on physiological parameter monitoring

Abstract

The invention discloses a traditional Chinese medicine external treatment control method based on physiological parameter monitoring, and relates to the technical field of traditional Chinese medicine external treatment control, and the method comprises the following steps: extracting a turning angle sequence and a direction jump sequence of a skin electric response signal and a heart rate variability signal in a trend same-frequency reference plane, constructing a trend dislocation tensor diagram, and obtaining a trend dislocation tensor diagram; according to the trend dislocation tensor diagram, judging whether trend dislocation exists between the skin electric response signal and the heart rate variability signal; and when the judgment result of the trend dislocation tensor diagram is that trend dislocation exists, continuous window analysis is carried out on the temperature signal, the change convergence rate, the mean translation rate and the amplitude saturation are calculated, and a thermostable focusing factor is generated. According to the method, the problem that the sub-stress state cannot be identified under the condition of physiological parameter trend dislocation is solved, accurate fusion judgment and dynamic external treatment regulation and control based on the trend dislocation tensor diagram and the thermal stability focusing factor are realized, and advanced identification and real-time intervention of the sub-stress state are completed.

Description

Technical Field

[0001] This invention relates to the field of TCM external treatment control technology, specifically to a TCM external treatment control method based on physiological parameter monitoring. Background Technology

[0002] Traditional Chinese medicine external treatment control based on physiological parameter monitoring refers to the use of sensor technology to collect key physiological parameters of the human body (such as heart rate, respiratory rate, body temperature, skin conductance, blood oxygen saturation, etc.) in real time or periodically, and to judge the individual's current physiological state through data processing and analysis. Based on this, the treatment method, duration, intensity and location of traditional Chinese medicine external treatment methods (such as moxibustion, hot compress, cupping, meridian electrical stimulation, etc.) are intelligently adjusted and controlled to achieve a personalized and dynamic treatment process. In existing technologies, such systems typically consist of five core components: First, physiological signal acquisition, primarily achieved through wearable devices (such as wristbands, chest straps, and skin patches); second, data transmission and storage, where the collected raw data is sent to a local processing terminal or cloud server via Bluetooth, Wi-Fi, or other means; third, data analysis and health assessment, where the system uses algorithmic models to identify anomalies and assess the status of physiological data, determining whether the user is in a state requiring treatment; fourth, intelligent decision-making for treatment plans, where the system generates personalized TCM external treatment control parameters based on TCM diagnostic theories and an expert rule base, combined with the current physiological state; and fifth, intelligent execution and feedback of external treatment devices, where the system issues control commands to moxibustion devices, hot compress devices, pulse electrical stimulators, etc., to execute corresponding treatments, and further optimizes the treatment strategy by continuously monitoring and providing feedback results. Overall, this technology integrates physiological monitoring, data communication, intelligent analysis, and TCM treatment control, representing a key path for the development of traditional TCM external treatment methods towards intelligence and precision.

[0003] The existing technology has the following shortcomings: In the process of using physiological parameter monitoring to control external TCM treatments, a special state arises when the trends of skin conductance response (SCR) and heart rate variability (HRV) signals show a slight temporal misalignment, while the temperature signal simultaneously exhibits a trend towards stabilization. In this situation, because SCR reflects sympathetic activity at the skin surface, HRV reflects overall autonomic nervous system regulation, and the temperature signal is related to peripheral microcirculation regulation, these signals undergo brief, rhythmic regulatory actions before entering a sub-stress state, naturally resulting in a trend misalignment. However, existing TCM external treatment control technologies based on physiological parameter monitoring cannot accurately identify whether an individual is in a sub-stress state based on the trend misalignment of SCR and HRV signals and the stabilization of temperature fluctuations. This is because current technologies generally rely on the synchronous changes between physiological parameters for judgment; once the three signals are out of sync, they are treated as invalid data or noise and ignored, making this early state completely unrecognizable. This can prevent the system from taking any TCM external treatment intervention in the initial stage of sub-stress, allowing the body to continue to develop to a higher level of activation. This not only misses the best time for conditioning, but also creates a blind spot in the recognition of such real physiological states over a long period of time. As a result, the entire TCM external treatment control logic based on physiological parameter monitoring gradually deviates from the real physiological changes, causing a series of serious effects such as decreased efficacy, distortion of the state model, and delayed response to external treatment.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method for controlling external treatments in traditional Chinese medicine based on physiological parameter monitoring, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a traditional Chinese medicine external treatment control method based on physiological parameter monitoring, specifically including the following steps: S1. Collect skin conductance response signals, heart rate variability signals and temperature signals, and generate a trend-frequency reference surface based on a unified time base, so that the collected skin conductance response signals, heart rate variability signals and temperature signals form a continuous trend trajectory within the trend-frequency reference surface; S2. Extract the turning angle sequence and direction jump sequence of the skin conductance response signal and heart rate variability signal within the trend-frequency reference plane, construct a trend misalignment tensor map, and determine whether there is a trend misalignment between the skin conductance response signal and the heart rate variability signal based on the trend misalignment tensor map. S3. When the trend misalignment tensor plot indicates that a trend misalignment exists, perform continuous window analysis on the temperature signal, calculate the convergence rate of change, the mean shift rate and the amplitude saturation, generate the thermal stability focusing factor, and determine whether the temperature signal exhibits fluctuation stabilization. S4. When the thermal stability focusing factor characterizes the temperature fluctuation to stabilize, the trend characteristics of the fusion trend misalignment tensor plot and the convergence characteristics of the thermal stability focusing factor are used to generate a sub-stress fusion sequence number, and the individual is identified as being in a sub-stress state based on the sub-stress fusion sequence number. S5. When the sub-stress fusion sequence number is identified as a sub-stress state, the sub-stress fusion sequence number is used as the driving variable to match the three-segment meridian treatment trajectory map, and the meridian treatment trajectory is dynamically adjusted according to the real-time changes of the trend misalignment tensor map and the thermal stability focusing factor, so as to complete the continuous execution of TCM external treatment control.

[0007] Preferably, S1 is as follows: The device continuously collects three types of physiological parameters—skin conductance response signal, heart rate variability signal, and temperature signal—using a time synchronization acquisition device, and adds a time stamp in a unified format to each type of physiological parameter, so that the skin conductance response signal, heart rate variability signal, and temperature signal have a unified time reference. Under a unified time reference, the skin conductance response signal, heart rate variability signal and temperature signal are aligned according to the time mark, and a trend co-frequency reference surface is constructed in the three-dimensional parameter space with the time reference as a reference, so that the skin conductance response signal, heart rate variability signal and temperature signal form a mappable trend data point set in the trend co-frequency reference surface; Within the trend reference plane, the trend data points of the skin conductance response signal, heart rate variability signal, and temperature signal are continuously interpolated and smoothed to form a continuous trend trajectory in the form of a time series.

[0008] Preferably, S2 specifically includes the following steps: S201. Within the trend reference plane, the continuous trend trajectories of the skin conductance response signal and the heart rate variability signal are divided into time segments, and the trajectory direction change angle is calculated in each time segment to generate the turning angle sequence of the skin conductance response signal and the turning angle sequence of the heart rate variability signal. Based on the change amplitude of the trajectory direction between adjacent time segments, the directional jump sequence of the skin conductance response signal and the directional jump sequence of the heart rate variability signal are formed. S202. Arrange the turning angle sequence of the skin conductance response signal and the turning angle sequence of the heart rate variability signal according to a unified time base, and combine them together with the directional jump sequence of the skin conductance response signal and the directional jump sequence of the heart rate variability signal to form a trend misalignment tensor unit, so that each trend misalignment tensor unit contains the angle change difference and directional jump difference within a time segment, and combine the trend misalignment tensor units in time order to form a trend misalignment tensor graph. S203. By comparing whether the difference in angle change and the difference in direction jump of the continuous trend misalignment tensor units in the trend misalignment tensor plot are continuously greater than the preset difference threshold, when the difference in angle change and the difference in direction jump are greater than the preset difference threshold in multiple continuous trend misalignment tensor units, it is determined that there is a trend misalignment between the skin conductance response signal and the heart rate variability signal.

[0009] Preferably, S203 is as follows: Select consecutively arranged trend misalignment tensor units in the trend misalignment tensor plot, and read the angle change difference and direction jump difference of each trend misalignment tensor unit in sequence, so that each trend misalignment tensor unit corresponds to an angle change difference value and a direction jump difference value, so as to form a continuously comparable difference sequence group. The angle change difference and the direction jump difference in the difference sequence group are compared with the preset difference threshold respectively, and the comparison results are marked according to the time order of the trend misalignment tensor unit, so that each trend misalignment tensor unit has a judgment mark for whether the angle change difference is greater than the preset difference threshold and whether the direction jump difference is greater than the preset difference threshold. When all the criteria for the angle change difference and the criteria for the direction jump difference are greater than the preset difference threshold in multiple consecutively arranged trend misalignment tensor units in the difference sequence group, it is determined that there is a trend misalignment between the skin conductance response signal and the heart rate variability signal.

[0010] Preferably, S3 specifically includes the following steps: S301. When the trend misalignment tensor plot indicates that a trend misalignment exists, extract the temperature signal sequence corresponding to the trend misalignment time period, and divide the temperature signal sequence into multiple equal-length continuous time windows, forming a temperature micro-change segment within each time window. S302. Calculate the convergence rate, mean translation rate, and amplitude saturation within each temperature micro-variation segment. The convergence rate is calculated by the reduction ratio of the distance between the maximum and minimum values ​​in the preceding and following temperature micro-variation segments. The mean translation rate is calculated by the average change of the adjacent temperature micro-variation segments. The amplitude saturation is calculated by the degree of amplitude convergence of the continuous temperature peak-valley fluctuations. S303. The convergence rate of change, the mean shift rate and the amplitude saturation are weighted and combined to generate a thermal stability focusing factor to characterize the stable trend of the temperature signal. It is determined whether the value of the thermal stability focusing factor is continuously within the preset value range. If multiple consecutive thermal stability focusing factors are within the preset value range, it is determined that the temperature signal has a fluctuating and stabilizing performance.

[0011] Preferably, S302 is as follows: Within each temperature micro-variation segment formed by continuous window analysis, all temperature data points of the segment are traversed, the maximum and minimum temperature values ​​in the segment are recorded, and the distance difference between the maximum and minimum temperature values ​​of adjacent temperature micro-variation segments is compared. This distance difference is used to calculate the change convergence rate, which reflects the degree of contraction in the temperature change amplitude of adjacent temperature micro-variation segments. Between adjacent temperature variation segments, the average temperature in each temperature variation segment is calculated, and the difference between the average temperature of adjacent temperature variation segments is calculated. This difference is used as the basis for calculating the mean translation rate. The mean translation rate reflects the speed of the overall temperature trend and is used to determine whether the temperature change tends to stabilize. Based on the temperature peak and valley data in continuous temperature micro-variation segments, the amplitude range of multiple continuous temperature peak and valley fluctuations is extracted, and the convergence degree between the continuous amplitude ranges is numerically quantified to form amplitude saturation. Amplitude saturation reflects whether the temperature fluctuation shows a stabilization trend, and together with the change convergence rate and mean shift rate, it is used to generate a comprehensive evaluation basis for temperature stability.

[0012] Preferably, S4 specifically includes the following steps: S401. When the thermal stability focusing factor characterizes the temperature fluctuation to stabilize, extract the angle change difference sequence and the direction jump difference sequence of the trend misalignment time period in the trend misalignment tensor. By numerically normalizing the angle change difference sequence and the direction jump difference sequence, obtain the trend extension level, trend offset density and trend misalignment intensity, which are used as trend features of the trend misalignment tensor. S402. Extract the continuous change sequence of the thermal stability focusing factor within the trend misalignment period. Calculate the convergence duration, convergence amplitude contraction rate, and convergence change span of the thermal stability focusing factor to form a convergence feature that characterizes the convergence change process of the temperature signal. Then, fuse this feature with the trend features of the trend misalignment tensor plot point by point according to the time sequence within the trend misalignment period to generate a fusion feature sequence of the trend misalignment tensor plot and the thermal stability focusing factor. S403. Encode each set of trend features and convergence features in the fusion feature sequence to form a fusion coding sequence that reflects the joint pattern of trend shift and temperature convergence. Arrange the fusion coding sequence in chronological order to form a sub-stress fusion sequence number. By comparing the coding arrangement structure of the sub-stress fusion sequence number with the preset sub-stress pattern sequence segment by segment, when the sub-stress fusion sequence number is consistent with the preset sub-stress pattern sequence in a continuous time period, identify the individual as being in a sub-stress state based on the sub-stress fusion sequence number.

[0013] Preferably, S402 is as follows: During the trend misalignment period, the continuous change values ​​of the thermal stability focusing factor are traversed in chronological order. The thermal stability focusing factor segments that are continuously and uninterruptedly in the stable range are marked. The convergence duration is calculated by statistically analyzing the start and end times of the segments. The convergence amplitude compression rate is calculated based on the maximum compression amplitude of the numerical fluctuations between adjacent thermal stability focusing factor segments. The convergence change span is calculated based on the maximum change range of the thermal stability focusing factor values ​​during the entire trend misalignment period. The convergence duration, convergence amplitude contraction rate, and convergence change span corresponding to each thermally stable focusing factor segment are combined into a convergence feature group corresponding to the time point. Then, according to each time point in the trend misalignment period, the convergence feature group is fused point by point with the trend extension level, trend offset density, and trend misalignment intensity at the same time point in the trend misalignment tensor to generate the fusion feature point of the trend misalignment tensor and the thermally stable focusing factor. The fusion feature points are sequentially spliced ​​together in chronological order to construct a fusion feature sequence of trend misalignment tensor and thermal stability focusing factor, which serves as the basis for generating subsequent sub-stress fusion sequence numbers.

[0014] Preferably, S5 is as follows: When the sub-stress fusion sequence number is identified as a sub-stress state, the encoded sequence in the sub-stress fusion sequence number is matched with the trajectory segment number in the three-segment meridian treatment trajectory diagram according to the time sequence. By comparing the sequence rhythm of the sub-stress fusion sequence number with the segment rhythm of the three-segment meridian treatment trajectory diagram, the three-segment meridian treatment trajectory diagram that matches the sub-stress fusion sequence number is determined, so that the segment arrangement of the meridian treatment trajectory diagram is matched by the sub-stress fusion sequence number as the driving variable. After completing the matching of the three-segment meridian treatment trajectory map, the trend extension level, trend offset density and trend misalignment intensity and the change convergence rate of the thermal stability focusing factor, mean translation rate and amplitude saturation in the trend misalignment tensor map are input into the matching results in time order. By mapping the trend characteristics of the trend misalignment tensor map to the convergence characteristics of the thermal stability focusing factor point by point, a dynamic control factor is formed to adjust the node parameters of the three-segment meridian treatment trajectory map. The dynamic regulation factor is mapped to each trajectory node of the three-segment meridian treatment trajectory map, so that the treatment intensity, treatment rhythm and treatment duration of the trajectory node are updated according to the real-time changes of the trend misalignment tensor map and the thermal stability focusing factor. Under the guidance of the sub-stress fusion sequence number, the meridian treatment trajectory is continuously regulated during the execution process to complete the continuous execution of TCM external treatment control.

[0015] The technical effects and advantages provided by the present invention in the above technical solution are as follows: 1. This invention, by introducing a trend misalignment tensor and a thermal stability focusing factor, achieves dynamic collaborative analysis of skin conductance response (SCR), heart rate variability (HRV), and temperature signals in asynchronous states, overcoming the limitation of reliance on signal synchronicity in traditional physiological parameter monitoring. When there is a trend shift between SCR and HRV signals, while the temperature signal simultaneously exhibits a fluctuating-to-stabilizing trend, this scheme can identify the trend misalignment characteristics by constructing a trend misalignment tensor and, combined with the convergence characteristics of the temperature signal within a local time period, generate a thermal stability focusing factor reflecting the actual physiological state changes. Furthermore, by temporally fusing the trend misalignment characteristics with the convergence characteristics of the thermal stability focusing factor, a sub-stress fusion sequence number is constructed, improving the accuracy of sub-stress state identification and early warning capabilities, filling the blind spot of existing technologies in asynchronous parameter situations.

[0016] 2. This invention uses the sub-stress fusion sequence number as the driving variable, matches a three-segment meridian treatment trajectory map, and constructs dynamic regulatory factors by combining a trend misalignment tensor map with real-time changes in thermal stability focusing factors. This allows for dynamic adjustment and closed-loop control of the TCM external treatment intervention path after identifying the sub-stress state, resulting in higher individual adaptability and timely response capabilities. By mapping the trend characteristics of multi-source physiological parameters and regulatory factors to each trajectory node of the meridian treatment trajectory map, the treatment rhythm, intensity, and duration are updated in real-time according to individual state fluctuations. This improves the response efficiency and regulatory precision of external treatment intervention in the sub-stress state, thereby enhancing overall efficacy and preventing further deterioration of the body due to recognition delays or lags in response. This strengthens the stability, accuracy, and practicality of the TCM external treatment control system based on physiological parameter monitoring. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0018] Figure 1 This is a flowchart illustrating the TCM external treatment control method based on physiological parameter monitoring according to the present invention. Detailed Implementation

[0019] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0020] This invention provides, for example Figure 1 The TCM external treatment method for controlling disease based on physiological parameter monitoring, as shown, specifically includes the following steps: S1. Collect skin conductance response signals, heart rate variability signals and temperature signals, and generate a trend-frequency reference surface based on a unified time base, so that the collected skin conductance response signals, heart rate variability signals and temperature signals form a continuous trend trajectory within the trend-frequency reference surface; In this embodiment, S1 specifically refers to: The device continuously collects three types of physiological parameters—skin conductance response signal, heart rate variability signal, and temperature signal—using a time synchronization acquisition device, and adds a time stamp in a unified format to each type of physiological parameter, so that the skin conductance response signal, heart rate variability signal, and temperature signal have a unified time reference. Skin conductance response signals, heart rate variability signals, and temperature signals originate from different physiological channels. During acquisition, timeline misalignment is highly likely due to sensor response delays, differences in communication protocols, and varying data processing queues. Therefore, continuous acquisition of these three types of signals using a time-synchronized acquisition device is crucial to ensuring data temporal consistency. Specifically, a central control chip generates a constant-frequency time reference signal. This time reference serves as a global timestamp source, broadcast to each signal acquisition node. Upon receiving physiological data, each node immediately appends a corresponding, uniformly formatted timestamp. Recording all signals on a unified timeline ensures comparability between different signal types at the same time point. This is crucial for subsequent trend-synchronous reference surface construction, trend feature extraction, and trend consistency analysis. This acquisition method not only eliminates time offset errors in various sensor signals but also provides a synchronous foundation for subsequent trend judgment and multi-parameter fusion.

[0021] A time-synchronized acquisition device refers to a signal acquisition and control unit with unified clock control capabilities, capable of coordinating multiple sensors to simultaneously acquire data and generate consistent time stamps. Skin conductance signals reflect the immediate impact of sympathetic nerve activity on skin conductance, typically acquired via micro-voltages from skin surface electrodes. Heart rate variability signals reflect the rhythm regulation of the autonomic nervous system, usually calculated based on differences in ECG intervals or pulse wave variations. Temperature signals primarily represent the microcirculatory heat distribution in peripheral or localized skin areas, commonly acquired using thermistors or miniature thermocouple sensors. A unified time stamp is a timestamp appended to each sampling point, its structure including a time value accurate to milliseconds and a sampling sequence number, ensuring strict data alignment in subsequent analysis. A unified time reference refers to a clock signal source or sampling reference commonly followed by various acquisition devices; it is a prerequisite for cross-channel fusion analysis of different signals. Without this unified time reference, different signals cannot be effectively synchronized in the time domain, resulting in inconsistent trend trajectories.

[0022] Under a unified time reference, the skin conductance response signal, heart rate variability signal and temperature signal are aligned according to the time mark, and a trend co-frequency reference surface is constructed in the three-dimensional parameter space with the time reference as a reference, so that the skin conductance response signal, heart rate variability signal and temperature signal form a mappable trend data point set in the trend co-frequency reference surface; After obtaining a unified time stamp, the skin conductance response (SCR) signal, heart rate variability (HRV) signal, and temperature signal are aligned point-by-point according to the time stamp. This alignment process requires interpolation and normalization of signals with different sampling frequencies to ensure that each time point corresponds to a three-parameter data point. Subsequently, using the time stamp as a unified horizontal axis, the SCR signal, HRV signal, and temperature signal are mapped onto three independent coordinate axes in a three-dimensional parameter space. With the time reference as the core, each three-parameter data point is positioned in the three-dimensional parameter space, thus constructing a trend-synchronous reference surface. This reference surface is formed by continuously connecting the spatial trajectories of the three types of signals at each time point, creating a set of trend data points. This approach transforms multiple physiological parameters of different dimensions into time-consistent multidimensional data trajectories, allowing observation of the interrelationships and trend synchronicity between parameters in a unified analysis space, laying the foundation for subsequent trend misalignment identification and abnormal state judgment.

[0023] The three-dimensional parameter space is a data analysis space composed of skin conductance response signal, heart rate variability signal, and temperature signal as the X, Y, and Z axes, respectively. In this space, each coordinate point represents the state of the three physiological parameters at a certain point in time. The trend-synchronous reference surface is a signal trajectory surface constructed in the three-dimensional parameter space based on time markers. This trajectory surface continuously connects each set of three parameter values ​​in chronological order, reflecting the mutual trend of each physiological signal within a unified time dimension. The trend data point set is the smallest unit constituting the trend-synchronous reference surface, consisting of the values ​​of skin conductance response signal, heart rate variability signal, and temperature signal at each moment. The continuity and spatial distribution of these trend data point sets are the basis for subsequent judgment of signal trend consistency, differences, and trend misalignment. Through this multi-parameter synchronous mapping method, the barrier between time domain and spatial characteristics can be broken down, enabling originally asynchronous and heterogeneous signals to have a unified analysis logic and path.

[0024] Within the trend reference plane, the trend data points of the skin conductance response signal, heart rate variability signal, and temperature signal are continuously interpolated and smoothed to form a continuous trend trajectory in the form of a time series.

[0025] Within the trend reference plane, to ensure that the skin conductance response signal, heart rate variability signal, and temperature signal exhibit continuous and identifiable trend paths, continuous interpolation and smoothing processing must be performed on the trend data point set constituting each type of signal. Continuous interpolation processing can be achieved through spline interpolation or local polynomial interpolation methods to fill the time intervals between adjacent data points, giving the signal trajectory fine-grained continuity on the time axis. Smoothing processing uses methods such as moving average or weighted exponential filtering to remove instantaneous fluctuations and high-frequency noise, making the signal more stable and representative in trend expression. After these two processes, the three types of signals are reconstructed into continuous trend trajectories with temporal consistency, coherent changes, and morphological resolvability. This processing method can enhance the readability and trend feature recognition capability of the signal without changing the essential nature of the original signal changes, and is a fundamental step for subsequent trend feature extraction and deviation analysis.

[0026] Continuous interpolation and smoothing are key techniques for transforming discrete signal sequences into sequences with continuous trends. Continuous interpolation fills in missing time points between existing sampling points, ensuring the signal's continuity over time and preventing trend breaks caused by different sampling frequencies or asynchronous sampling moments. Smoothing reduces the impact of short-term fluctuations in the original signal, extracting its core trend pattern within a time window to highlight medium- to long-term trend characteristics. Continuous trend trajectories refer to the continuous curves formed after the above processing, where each type of signal is arranged chronologically within a trend-frequency reference plane. These curves reflect the true path of the physiological parameter's change over time. These continuous trend trajectories not only visually present the signal's change pattern but also provide trajectory-level data support for subsequent trend misalignment judgment and state identification.

[0027] S2. Extract the turning angle sequence and direction jump sequence of the skin conductance response signal and heart rate variability signal within the trend-frequency reference plane, construct a trend misalignment tensor map, and determine whether there is a trend misalignment between the skin conductance response signal and the heart rate variability signal based on the trend misalignment tensor map. In this embodiment, S2 specifically includes the following steps: S201. Within the trend reference plane, the continuous trend trajectories of the skin conductance response signal and the heart rate variability signal are divided into time segments, and the trajectory direction change angle is calculated in each time segment to generate the turning angle sequence of the skin conductance response signal and the turning angle sequence of the heart rate variability signal. Based on the change amplitude of the trajectory direction between adjacent time segments, the directional jump sequence of the skin conductance response signal and the directional jump sequence of the heart rate variability signal are formed. Within the trend-frequency reference plane, the continuous trend trajectories of the EEG and heart rate variability signals are uniformly divided based on a unified time base, with each segment constituting a fixed duration. Within each time segment, the first and last two trend data points of the trajectory are extracted, and the direction vector of the trajectory is calculated based on their spatial coordinates within the trend-frequency reference plane. The angle between the direction vectors of adjacent time segments is then calculated to obtain the trajectory direction change angle of the current segment. Repeating this process generates a set of time-ordered turning angle sequences for both the EEG and heart rate variability signals. The magnitude of the directional jump is then obtained by calculating the numerical difference between adjacent turning angles in these sequences, thus forming a directional jump sequence. For example, for the EEG signal, if the turning angles of two consecutive time segments are 30 degrees and 50 degrees, the directional jump is 20 degrees. This processing method effectively captures abrupt changes in the direction of the trend trajectory at the microscopic level, helping to identify "asynchronous disturbances" in the signal rhythm, which is a typical manifestation of trend misalignment.

[0028] The angle of change in trajectory direction refers to the angle formed by the change in direction of a signal's trend path relative to the previous segment within a certain time segment. It is a direct quantitative form of the intensity of the change in trend trajectory direction. The turning angle sequences of skin conductance response (SCR) signals and heart rate variability (HRV) signals are sets of the angles of change in direction for these two types of signals in continuous time segments, respectively, demonstrating the rhythmic adjustment process of their trend trajectories in the time dimension. The amplitude of the change in trajectory direction between adjacent time segments is the angle jump value obtained by the difference in turning angles between the two time segments, used to reveal the abrupt change characteristics of the signal during trend transition. The directional jump sequences of SCR signals and HRV signals are time series sets calculated by the difference between adjacent values ​​of the turning angle sequences, used to characterize the volatility and instability of the signal in the trend trajectory, thus providing a highly sensitive indicator for subsequent trend misalignment identification. The construction of these sequences not only realizes the conversion from the time domain to the angle domain but also provides multi-dimensional feature support for trend comparison.

[0029] S202. Arrange the turning angle sequence of the skin conductance response signal and the turning angle sequence of the heart rate variability signal according to a unified time base, and combine them together with the directional jump sequence of the skin conductance response signal and the directional jump sequence of the heart rate variability signal to form a trend misalignment tensor unit, so that each trend misalignment tensor unit contains the angle change difference and directional jump difference within a time segment, and combine the trend misalignment tensor units in time order to form a trend misalignment tensor graph. To achieve a structured representation of trend misalignment features, the turning angle sequences of the EEG and heart rate variability signals are first arranged in time segments according to a unified time base. This ensures that every data point in both angle sequences falls within the same time window and maintains a strict one-to-one correspondence within the sequences. Subsequently, the difference between the turning angles of the EEG and heart rate variability signals is calculated in each time segment to obtain the angle change difference. The difference between the directional jump values ​​of the EEG and heart rate variability signals is then calculated to obtain the directional jump difference. These angle change differences and directional jump differences together constitute a multidimensional data unit, namely, a trend misalignment tensor unit. Each tensor unit reflects the degree of asynchrony between the two signals in terms of trend direction and fluctuation amplitude within a time segment. By concatenating the trend misalignment tensor units generated in all time segments in chronological order, a trend misalignment tensor map can be constructed. Taking a real-world scenario as an example, if the turning angle of the skin conductance response signal is 25 degrees, the heart rate variability signal is 15 degrees, and the directional jumps are 10 degrees and 6 degrees respectively within a certain time segment, then the tensor unit of that time segment contains an angle difference of 10 degrees and a jump difference of 4 degrees. This method helps to capture the persistent distribution of inconsistencies between trends on the time axis.

[0030] Arranging signals according to a unified time base means synchronizing all signal parameters on the same time baseline, ensuring strict temporal consistency of data points in each time segment. This is the foundation for ensuring the effectiveness of comparisons between tensor units. A trend misalignment tensor unit is a data unit containing two-dimensional differences, specifically used to express the degree of trend asynchrony between skin conductance response signals and heart rate variability signals within a single time window. The angle change difference reflects the degree of separation in trend direction, while the directional jump difference reflects the difference in the amplitude of short-term trend fluctuations. The angle change difference is the absolute numerical difference between the turning angles of two signals within a unified time segment, representing the inconsistency in trend direction. The directional jump difference is a quantitative expression of the difference in the severity of trend abrupt changes between the two signals within that segment. The trend misalignment tensor plot is a time series graph composed of a series of trend misalignment tensor units arranged in chronological order. It can reveal the onset, duration, and decay of trend misalignment states in the time dimension, serving as the core basis for subsequent judgment of the existence and persistence of trend misalignment. This graph structure not only improves the spatial sensitivity of signal processing but also enhances the traceability of dynamic trend evolution.

[0031] S203. By comparing whether the difference in angle change and the difference in direction jump of the continuous trend misalignment tensor units in the trend misalignment tensor plot are continuously greater than the preset difference threshold, when the difference in angle change and the difference in direction jump are greater than the preset difference threshold in multiple continuous trend misalignment tensor units, it is determined that there is a trend misalignment between the skin conductance response signal and the heart rate variability signal.

[0032] By comparing the differences in angle changes and directional jumps between consecutive trend misalignment tensor units in the trend misalignment tensor plot to ensure they consistently exceed a preset threshold, the aim is to identify whether stable and significant inconsistent regulatory trends exist between skin conductance response (SCR) signals and heart rate variability (HRV) signals within a specific time range. SCR signals primarily reflect transient fluctuations in peripheral sympathetic activity, while HRV signals represent the overall regulatory rhythm of the autonomic nervous system; both typically exhibit high synchronicity under normal conditions. Once their trends show a stable temporal misalignment, especially exhibiting sustained deviations in direction and angle and a continuous increase in fluctuation jump amplitude across multiple time segments, it is highly likely an early sign that the individual's neuro-blood flow regulation mechanism has entered a sub-stress transitional state. Identification using this method effectively distinguishes trend disturbances caused by occasional noise from genuine functional disorder trends, thus providing a more targeted basis for subsequent interventions. This design possesses high robustness and sensitivity, avoiding misjudgments caused by short-term fluctuations and helping to detect potential abnormal states early.

[0033] In this embodiment, S203 specifically refers to: Select consecutively arranged trend misalignment tensor units in the trend misalignment tensor plot, and read the angle change difference and direction jump difference of each trend misalignment tensor unit in sequence, so that each trend misalignment tensor unit corresponds to an angle change difference value and a direction jump difference value, so as to form a continuously comparable difference sequence group. To quantitatively identify trend misalignment, a continuous sequence of trend misalignment tensor cells needs to be selected from the trend misalignment tensor graph. The angle change difference and directional jump difference within each tensor cell are then read sequentially. The two difference values ​​from each time segment are extracted chronologically, constructing two independent but synchronized numerical sequences. Specifically, a sliding window approach can be used to extract a certain number of continuous tensor cells from the trend misalignment tensor graph and record the two difference values ​​in each cell, thus forming two time-corresponding difference sequence sets. These two difference sequences represent the dynamic differences in trend direction and short-term fluctuations of skin conductance response signals and heart rate variability signals within consecutive time segments. The core value of this difference sequence set lies in its continuous comparability, allowing for cross-segment horizontal comparisons along the time axis to analyze the changing patterns of the differences over time. This method not only detects trend deviations within a single time segment but also captures whether trend deviations across multiple time segments exhibit persistence or accumulation, thus providing data support for accurate trend misalignment assessment. Such difference sequences can also form a dynamic map of trend misalignment in visual presentation, which facilitates subsequent identification and intervention strategy formulation.

[0034] The angle change difference and the direction jump difference in the difference sequence group are compared with the preset difference threshold respectively, and the comparison results are marked according to the time order of the trend misalignment tensor unit, so that each trend misalignment tensor unit has a judgment mark for whether the angle change difference is greater than the preset difference threshold and whether the direction jump difference is greater than the preset difference threshold. After constructing the difference sequence group, each angle change difference and directional jump difference in the sequence needs to be compared item by item with a preset difference threshold. The comparison can be done using simple numerical logic operations, such as determining a greater than / less than relationship between each pair of differences and the corresponding threshold, resulting in a Boolean value. The results of each comparison need to be marked sequentially according to the time order of the trend misalignment tensor units, forming a set of judgment mark sequences. This ensures that each trend misalignment tensor unit corresponds to both an angle change difference judgment mark and a directional jump difference judgment mark. This process can be implemented using a sliding window combined with an item-by-item threshold judgment function, storing the judgment results in an array in chronological order each time the window slides. The preset difference threshold is typically set based on historical large-sample statistical analysis, extracting the mean and variance of differences under stable trend conditions from real case data, and combining this with sensitivity analysis to select a numerical range that reflects the critical point of trend misalignment. The threshold for angle change differences is used to determine whether abrupt changes in trajectory direction are significant, and the threshold for directional jump differences is used to determine whether short-term directional fluctuations are abnormal. The combination of these two allows for a dual determination of trend misalignment trends. The establishment of this time series labeling mechanism lays a structured data foundation for subsequent detection of the persistence of trend misalignment across multiple consecutive time segments.

[0035] When all the criteria for the angle change difference and the criteria for the direction jump difference are greater than the preset difference threshold in multiple consecutively arranged trend misalignment tensor units in the difference sequence group, it is determined that there is a trend misalignment between the skin conductance response signal and the heart rate variability signal, and this trend misalignment conclusion is used as the input condition for subsequent trend analysis.

[0036] In the process of trend misalignment judgment, by continuously reading the judgment markers of angle change difference and direction jump difference, a set of judgment results arranged in sequence can be formed in the time dimension. Within the continuously arranged multiple trend misalignment tensor units in the difference sequence group, it refers to a continuous time segment connected in chronological order in the trend misalignment tensor graph, where each time segment corresponds to a trend misalignment tensor unit, and these units cannot be interrupted or jumped. Within this continuous unit, the corresponding angle change difference judgment markers and direction jump difference judgment markers are logically judged. When all markers are greater than a preset difference threshold, the trend misalignment phenomenon can be considered to exist stably within this time interval. This judgment can be implemented using sliding window logic, selecting a continuous unit sequence of fixed length for overall judgment each time. If all conditions are met within the sliding window, the trend misalignment is determined to be valid. For example, in a sliding window of length five, if the angle difference and direction difference judgments of all five units are greater than the threshold, a trend misalignment signal is output. This judgment mechanism ensures that the trend misalignment phenomenon has sufficient persistence and significance, eliminates the interference of occasional fluctuations on the judgment results, and at the same time, the conclusion of trend misalignment will be used as the premise for subsequent analysis, ensuring that subsequent judgments are targeted and timely.

[0037] S3. When the trend misalignment tensor plot indicates that a trend misalignment exists, perform continuous window analysis on the temperature signal, calculate the convergence rate of change, the mean shift rate and the amplitude saturation, generate the thermal stability focusing factor, and determine whether the temperature signal exhibits fluctuation stabilization. In this embodiment, S3 specifically includes the following steps: S301. When the trend misalignment tensor plot indicates that a trend misalignment exists, extract the temperature signal sequence corresponding to the trend misalignment time period, and divide the temperature signal sequence into multiple equal-length continuous time windows, forming a temperature micro-change segment within each time window. Once the trend misalignment tensor plot indicates the presence of a trend misalignment, the first step is to determine the start and end time range corresponding to the trend misalignment within the trend-frequency reference plane. Using this time range as an index, a complete temperature signal sequence is extracted from the continuous temperature signal. This temperature signal sequence must maintain a time resolution consistent with the skin conductance response signal and heart rate variability signal to ensure accurate time correspondence. The extracted temperature signal sequence is divided into several consecutive, non-overlapping time windows of a fixed length. Each time window can be 10 seconds, 20 seconds, or other conventional time units selected based on physiological response characteristics. Within each time window, all temperature sampling points are extracted in chronological order, forming a temperature micro-variation segment containing multiple continuous temperature change data. This operation can be implemented using a sliding window algorithm, achieving equal-length, continuous division under the condition that the window length and step size are equal. The key to this processing flow is compressing the continuity of the temperature signal into multiple micro-variation segments that can be processed in parallel, allowing for independent calculation of the trend characteristic values ​​within each segment.

[0038] The temperature signal sequence corresponding to the trend misalignment period refers to the temperature signal synchronously extracted according to the time consistency rule within the time interval where the trend misalignment between the skin conductance response signal and the heart rate variability signal is detected. This temperature signal sequence retains all the dynamic information of the temperature during this abnormal adjustment period. Multiple equal-length continuous time windows refer to the non-overlapping segmentation structure of the temperature signal sequence in chronological order, using a uniform time length as the unit, ensuring that the temperature data in each window covers the same time range, facilitating the consistency of feature calculation. Temperature micro-variation segments refer to time series segments composed of multiple continuous temperature data points formed within each time window, used to characterize the short-term dynamic trend of temperature changes within that window. Temperature micro-variation segments serve as the basic unit for subsequent calculations of convergence rate, mean shift rate, and amplitude saturation; their structural definition and segmentation accuracy directly affect the accuracy of fluctuation stabilization determination. Through this processing method, which uses the trend misalignment period as an index, continuous window division as the structure, and micro-variation segments as the analysis unit, high-precision modeling of the stability characteristics of temperature signals can be achieved.

[0039] S302. Calculate the convergence rate, mean translation rate, and amplitude saturation within each temperature micro-variation segment. The convergence rate is calculated by the reduction ratio of the distance between the maximum and minimum values ​​in the preceding and following temperature micro-variation segments. The mean translation rate is calculated by the average change of the adjacent temperature micro-variation segments. The amplitude saturation is calculated by the degree of amplitude convergence of the continuous temperature peak-valley fluctuations. S303. The convergence rate of change, the mean shift rate and the amplitude saturation are weighted and combined to generate a thermal stability focusing factor to characterize the stable trend of the temperature signal. It is determined whether the value of the thermal stability focusing factor is continuously within the preset value range. If multiple consecutive thermal stability focusing factors are within the preset value range, it is determined that the temperature signal has a fluctuating and stabilizing performance.

[0040] To accurately identify whether a temperature signal exhibits a fluctuating but stabilizing trend, a weighted combination of the previously calculated convergence rate, mean shift rate, and amplitude saturation is required. In practice, these three parameters are first standardized to ensure comparability on a numerical scale. Then, based on experience or statistical analysis, the weight of each parameter in characterizing temperature stability is determined; for example, the convergence rate is given a higher weight to highlight the trend of fluctuating contraction. The weighted value is the thermal stability focusing factor, which comprehensively reflects the stability characteristics of the temperature signal. Subsequently, using the thermal stability focusing factors generated in multiple consecutive time windows as a sequence, each factor is compared with a preset stability value range. If the thermal stability focusing factor falls within this preset range in multiple consecutive windows, it can be determined that the temperature signal has transitioned from a fluctuating phase to a stabilizing phase during that time period. This determination not only eliminates instantaneous fluctuations but also improves the continuity and reliability of the stability trend assessment.

[0041] Weighted combination is the process of constructing a unified evaluation index by fusing multiple independent indicators reflecting the fluctuation characteristics of temperature signals. The convergence rate of change, the mean shift rate, and the amplitude saturation represent different aspects of fluctuation contraction, overall trend shift, and amplitude compression, respectively. These three are integrated with fixed weights to form the thermal stability focusing factor. The thermal stability focusing factor is a comprehensive numerical indicator used to describe the stable trend of temperature signals; the more concentrated its value and the closer it is to the central range, the more stable the temperature signal. The preset numerical range is a stable interval set based on statistical analysis of a large amount of physiological data. This interval is used to define whether an individual is in a state of microcirculatory stabilization. By comparing the thermal stability focusing factor over multiple consecutive time periods with the preset numerical range, it is possible to dynamically determine whether an individual has experienced a stable transition in temperature signals, thus serving as an important basis for subsequent identification of sub-stress states.

[0042] In this embodiment, S302 specifically refers to: Within each temperature micro-variation segment formed by continuous window analysis, all temperature data points of the segment are traversed, the maximum and minimum temperature values ​​in the segment are recorded, and the distance difference between the maximum and minimum temperature values ​​of adjacent temperature micro-variation segments is compared. This distance difference is used to calculate the change convergence rate, which reflects the degree of contraction in the temperature change amplitude of adjacent temperature micro-variation segments. In continuous window analysis, each temperature micro-variation segment consists of temperature sampling points within a fixed time period. To analyze its fluctuation amplitude, all temperature sampling points within each segment need to be read and compared one by one to determine the maximum and minimum temperature values ​​within that segment. The maximum temperature value represents the upper temperature limit within the segment, and the minimum temperature value represents the lower temperature limit; the difference between the two is the temperature fluctuation range of that segment. For any two adjacent temperature micro-variation segments at any given time, their respective temperature fluctuation ranges are calculated, and the difference between these two ranges is quantified to calculate the degree of contraction in temperature amplitude between the preceding and following segments. When the fluctuation range of the later segment is smaller than that of the earlier segment, it indicates that the temperature amplitude is converging; this degree of convergence is the change convergence rate. A positive change convergence rate for multiple consecutive segments, along with satisfying the minimum contraction ratio, can be set as a criterion to determine whether the temperature is gradually approaching a stable trend. This processing logic is suitable for capturing the dynamic characteristics of the transition from fluctuation to steady state in micro-circulation regulation, and is particularly suitable for distinguishing temperature response states under asynchronous trends. The convergence rate, as a key component of the thermal stability focusing factor, quantifies whether temperature fluctuations are converging rapidly or slowly, providing a crucial indicator for ultimately determining whether the temperature signal exhibits a trend towards stability. The extraction of the maximum and minimum temperature values ​​requires traversing the entire temperature segment data to ensure the accuracy and stability of amplitude identification.

[0043] Between adjacent temperature variation segments, the average temperature in each temperature variation segment is calculated, and the difference between the average temperature of adjacent temperature variation segments is calculated. This difference is used as the basis for calculating the mean translation rate. The mean translation rate reflects the speed of the overall temperature trend and is used to determine whether the temperature change tends to stabilize. In multiple continuous temperature fluctuation segments, to characterize the overall trend of temperature signal changes, it is necessary to traverse all temperature data points in each segment and calculate its average temperature. The average temperature represents the overall thermal level within that time segment. Subsequently, the difference between the average values ​​of two adjacent temperature fluctuation segments is calculated in chronological order; the resulting value is the mean shift rate. This rate reflects whether the overall trend of the temperature signal is rising, falling, or remaining stable. When the mean shift rate of multiple adjacent temperature fluctuation segments approaches zero, it indicates that the rate of temperature change has slowed down and is approaching a stable state; conversely, it indicates that the temperature trend has not yet converged. To improve the accuracy of judgment, a threshold for the mean shift rate can be set to identify whether a stable state has been reached. For example, if the difference between the average values ​​of three consecutive fluctuation segments is less than a certain set value, it can be considered that the temperature change has tended to stabilize. As a parameter reflecting the rate of change of trend, the mean shift rate is of great significance in identifying the transition process from fluctuation to steady state in micro-circulation regulation. This parameter can further enhance the accuracy of judging whether the temperature signal shows a fluctuating trend towards stability. It is especially suitable for use in conjunction with the change convergence rate and amplitude saturation to improve the stability identification capability of multidimensional physiological trend fusion analysis.

[0044] Based on the temperature peak and valley data in continuous temperature micro-variation segments, the amplitude range of multiple continuous temperature peak and valley fluctuations is extracted, and the convergence degree between the continuous amplitude ranges is numerically quantified to form amplitude saturation. Amplitude saturation reflects whether the temperature fluctuation shows a stabilization trend, and together with the change convergence rate and mean shift rate, it is used to generate a comprehensive evaluation basis for temperature stability.

[0045] To determine whether a temperature signal has entered a stabilizing phase under a trend of misalignment, it is necessary to identify temperature peaks and troughs from continuous small temperature variation segments and calculate the amplitude range between each peak-trough pair, i.e., the numerical difference between the peak and trough values. The amplitude ranges in multiple continuous small temperature variation segments constitute an amplitude sequence of temperature fluctuations. This amplitude sequence is compared one by one to determine whether its changes show a gradually decreasing trend. If the differences between multiple consecutive amplitude ranges gradually converge to a smaller interval, it indicates that the temperature fluctuation amplitude is approaching saturation and the temperature state is gradually stabilizing. This convergence trend is quantified numerically using the difference contraction ratio or standard deviation volatility to form amplitude saturation. Amplitude saturation is used to quantify the convergence strength of continuous fluctuations and is an important technical indicator for identifying the stabilizing state of fluctuations. When the amplitude saturation value is low and the change is stable, it indicates that the temperature signal no longer exhibits drastic fluctuations, consistent with the pattern of microcirculation transitioning from the adjustment period to the stable period. Amplitude saturation, along with the convergence rate of change and the mean shift rate, constitute the core parameter group for multidimensional assessment of temperature stability. This group has high reliability and sensitivity, and helps to accurately determine whether an individual has entered the early critical point of sub-stress state in the context of trend misalignment.

[0046] S4. When the thermal stability focusing factor characterizes the temperature fluctuation to stabilize, the trend characteristics of the fusion trend misalignment tensor plot and the convergence characteristics of the thermal stability focusing factor are used to generate a sub-stress fusion sequence number, and the individual is identified as being in a sub-stress state based on the sub-stress fusion sequence number. In this embodiment, S4 specifically includes the following steps: S401. When the thermal stability focusing factor characterizes the temperature fluctuation to stabilize, extract the angle change difference sequence and the direction jump difference sequence of the trend misalignment time period in the trend misalignment tensor. By numerically normalizing the angle change difference sequence and the direction jump difference sequence, obtain the trend extension level, trend offset density and trend misalignment intensity, which are used as the trend features of the trend misalignment tensor. The trend features of the trend misalignment tensor are used to characterize the trend offset pattern of skin conductance response signal and heart rate variability signal within the trend misalignment time period. After the temperature signal is identified as exhibiting fluctuating stability through the thermal stability focusing factor, it is necessary to extract the trend misalignment time period synchronized with this process from the trend misalignment tensor graph. Within this time period, the angle change difference sequence and directional jump difference sequence of the skin conductance response signal and heart rate variability signal are obtained. Specifically, each trend misalignment tensor cell in the trend misalignment tensor graph can be traversed chronologically, and the angle change difference and directional jump difference within each time cell can be extracted as a set of trend data points. Then, multiple trend data points within the entire trend misalignment time period are numerically normalized to unify the dimensions, standardize the intervals, and remove noise points, resulting in three indicators for quantifying the trend pattern: the trend extension level can be calculated by measuring the distribution span of trend data points on the time axis, representing the range of trend shift duration; the trend shift density can be calculated by the number of data points experiencing trend abrupt changes per unit time, reflecting the concentration of trend disturbances; and the trend misalignment intensity can be measured by the combined amplitude average of the angle change difference and the directional jump difference, describing the overall intensity level of the trend misalignment. These features constitute the set of trend characteristics of the trend misalignment tensor plot within the trend misalignment period, laying the foundation for subsequent fusion analysis.

[0047] The trend misalignment time period refers to the trend misalignment tensor image segment that coincides with the time after the thermal stability focusing factor enters the stable interval, usually obtained through time stamp synchronization mapping. The angle change difference sequence is a sequence composed of the differences between the angles between the direction vectors of the skin conductance response signal and the heart rate variability signal within a continuous time window, used to characterize the dynamics of trend reversal; the direction jump difference sequence is used to express the degree of abrupt change in trend direction in adjacent time segments, reflecting the severity of trend jump behavior. Numerical normalization transforms data from different dimensions into a unified scale through standardization, normalization, or exponential compression, facilitating subsequent statistical analysis. The trend extension level reflects the persistence of trend changes over time, the trend offset density measures the spatiotemporal clustering of trend disturbances, and the trend misalignment intensity quantitatively characterizes the magnitude of trend deviation. The combination of these three forms a complete description of the trend offset pattern, enabling the trend misalignment tensor image to provide structured and numerical trend input for sub-stress fusion sequences.

[0048] S402. Extract the continuous change sequence of the thermal stability focusing factor within the trend misalignment period. Calculate the convergence duration, convergence amplitude contraction rate, and convergence change span of the thermal stability focusing factor to form a convergence feature that characterizes the convergence change process of the temperature signal. Then, fuse this feature with the trend features of the trend misalignment tensor plot point by point according to the time sequence within the trend misalignment period to generate a fusion feature sequence of the trend misalignment tensor plot and the thermal stability focusing factor. S403. Encode each set of trend features and convergence features in the fusion feature sequence to form a fusion coding sequence that reflects the joint pattern of trend shift and temperature convergence. Arrange the fusion coding sequence in chronological order to form a sub-stress fusion sequence number. By comparing the coding arrangement structure of the sub-stress fusion sequence number with the preset sub-stress pattern sequence segment by segment, when the sub-stress fusion sequence number is consistent with the preset sub-stress pattern sequence in a continuous time period, identify the individual as being in a sub-stress state based on the sub-stress fusion sequence number.

[0049] To identify an individual's sub-stress state, the time series data resulting from the fusion of trend and convergence features needs to be encoded. Specifically, each fused feature point's feature combination is first discretized and encoded. The trend extension level, trend offset density, trend misalignment intensity, convergence duration, convergence amplitude contraction rate, and convergence change span are mapped to fixed-length symbols or digital codes according to predefined numerical ranges. Then, the encoded combinations at each moment are concatenated into a continuous encoded string, forming a complete fused encoded sequence. Next, fixed-length segments are extracted from the fused encoded sequence using a sliding window approach and compared sequentially with an established sub-stress pattern sequence for similarity. If multiple consecutive segments in the fused encoded sequence match a preset sub-stress pattern sequence threshold within a certain time interval, the individual is determined to be in a sub-stress state based on the comparison results. This method effectively combines the synergistic characteristics of physiological trend changes and temperature stability signals to construct an accurate mechanism for discriminating an individual's psychological stress state.

[0050] The fusion-coded sequence reflecting the joint pattern of trend deviation and temperature convergence is obtained by transforming the trend characteristics of the trend misalignment tensor and the convergence characteristics of the thermal stability focusing factor into discrete coded fragments through numerical mapping. Each coded fragment represents the comprehensive state of an individual at a certain point in time across multiple physiological dimensions. The sub-stress fusion sequence number is the main coded sequence assembled from these fusion coded fragments in chronological order, presenting a structured sequence stream that can be used for comparison and identification. The preset sub-stress pattern sequence is a representative coded template constructed based on existing physiological experiments or individual historical data, reflecting the common trend misalignment and temperature convergence co-change patterns before an individual enters a sub-stress state. By comparing the real-time generated sub-stress fusion sequence number with the preset template sequence segment by segment, the presence of typical sub-stress physiological characteristic patterns can be identified from the joint trend of multi-dimensional dynamic changes, thereby achieving dynamic judgment of the individual's state.

[0051] In this embodiment, S402 specifically refers to: During the trend misalignment period, the continuous change values ​​of the thermal stability focusing factor are traversed in chronological order. The thermal stability focusing factor segments that are continuously and uninterruptedly in the stable range are marked. The convergence duration is calculated by statistically analyzing the start and end times of the segments. The convergence amplitude compression rate is calculated based on the maximum compression amplitude of the numerical fluctuations between adjacent thermal stability focusing factor segments. The convergence change span is calculated based on the maximum change range of the thermal stability focusing factor values ​​during the entire trend misalignment period. To accurately characterize the stabilization trend of the temperature signal during periods of trend misalignment, it is necessary to sequentially traverse the continuous changes in the thermal stability focusing factor. First, by setting a stable interval threshold range, a sequence of thermal stability focusing factor values ​​continuously falling within this range is extracted as a continuous, uninterrupted segment of the thermal stability focusing factor within the stable interval. For example, when the value of the thermal stability focusing factor is continuously between 0.4 and 0.6 for a duration exceeding a set minimum time threshold, it can be considered a valid segment. After obtaining multiple such segments, the convergence duration can be determined by calculating the start and end times of each segment. This indicator reflects the continuous maintenance capability of the thermal stability focusing factor within the stable interval. Subsequently, the compression amplitude of the numerical changes between two adjacent thermal stability focusing factor segments is recorded, i.e., the degree of reduction in the difference between the maximum and minimum values ​​between adjacent segments. This is used to calculate the convergence amplitude compression rate, which reflects the compression trend of the temperature signal during its convergence from a fluctuating state to a stable state. Finally, during the entire period of trend misalignment, the maximum fluctuation range of the thermal stability focusing factor is statistically analyzed, and combined with its convergence trend before and after, the convergence change span is calculated. This indicator is used to measure the total amplitude change during the process from fluctuation to stability. These three indicators work together to form a complete description of the convergence behavior of the temperature signal.

[0052] A continuous, uninterrupted segment of the thermally stable focusing factor within a stable range refers to a value of the thermally stable focusing factor falling within the stable range at multiple time points without interruption, demonstrating the consistent stability of the temperature signal. The convergence duration is the length of each stable segment, used to assess the ability to maintain a stable state. The convergence amplitude contraction rate is the degree to which the numerical difference between two adjacent segments decreases, measuring the decreasing trend of signal fluctuation intensity. The convergence change span is the largest overall change range during the process from instability to stability, reflecting the overall degree of stabilization of the temperature signal. These characteristics exhibit quantifiable evolutionary properties in the time series of the thermally stable focusing factor, providing convergence evidence in the temperature change dimension for sub-stress identification.

[0053] The convergence duration, convergence amplitude contraction rate, and convergence change span corresponding to each thermally stable focusing factor segment are combined into a convergence feature group corresponding to the time point. Then, according to each time point in the trend misalignment period, the convergence feature group is fused point by point with the trend extension level, trend offset density, and trend misalignment intensity at the same time point in the trend misalignment tensor to generate the fusion feature point of the trend misalignment tensor and the thermally stable focusing factor. To achieve accurate identification of sub-stress states, it is necessary to fuse convergence features reflecting temperature stability with trend features reflecting changes in skin conductance and heart rate variability signals point by point. First, for each time point within a trend misalignment period, three key quantitative indicators of the corresponding thermal stability focusing factor segment are extracted: convergence duration, convergence amplitude contraction rate, and convergence change span. These three indicators are then packaged into a complete convergence feature group at each time point. Subsequently, at the same time point, the corresponding trend extension level, trend offset density, and trend misalignment intensity are extracted from the trend misalignment tensor. The trend extension level represents the duration of trend misalignment, the trend offset density refers to the density of trend changes per unit time, and the trend misalignment intensity reflects the degree or severity of trend misalignment. By concatenating the convergence feature group at a time point with the trend features at that time point, a fused multidimensional vector is formed—the fused feature point of the trend misalignment tensor and the thermal stability focusing factor. Each fused feature point carries complete change information in both trend and temperature dimensions, achieving cross-coupling at the data level.

[0054] The convergence feature set consists of three indicators: convergence duration, convergence amplitude contraction rate, and convergence change span. These indicators characterize whether the temperature signal is converging from an unstable state to a stable state within a given time period. Trend extension level, trend offset density, and trend misalignment intensity are three trend quantification parameters calculated based on skin conductance response signals and heart rate variability signals. These parameters reflect the coupling offset state between signals from the autonomic nervous system. Point-by-point fusion refers to pairing and merging the convergence feature set with the trend feature at each time point within the trend misalignment time period, according to the principle of time point alignment, to form a time-corresponding fusion feature point sequence. The fusion feature points of the trend misalignment tensor map and the thermal stability focusing factor not only retain the change information of their respective signals but also provide the dynamic interaction relationship between them over time, thus providing a high-precision input basis for the subsequent generation of sub-stress fusion sequence numbers.

[0055] The fusion feature points are sequentially spliced ​​together in chronological order to construct a fusion feature sequence of trend misalignment tensor and thermal stability focusing factor, which serves as the basis for generating subsequent sub-stress fusion sequence numbers.

[0056] To achieve continuous identification of individual sub-stress states, it is necessary to sequentially concatenate multiple fusion feature points obtained within the trend misalignment period. Specifically, the timestamps of each fusion feature point are first sorted to ensure they are arranged in chronological order. Then, these fusion feature points are sequentially combined into a complete time-series vector set, constructing a fusion feature sequence of the trend misalignment tensor and the thermal stability focusing factor. Each unit in this sequence contains the trend extension level, trend offset density, trend misalignment intensity, and corresponding convergence duration, convergence amplitude contraction rate, and convergence change span at a given time point, ensuring that each fusion unit fully expresses the joint dynamic characteristics of trend and temperature changes. The data structure can be matrix-based, with each row representing a fusion time point and each column representing a feature dimension. This method of constructing the fusion feature sequence not only achieves continuity in the time dimension of the data but also provides accurate, structured, and quantifiable basic information input for the subsequent generation of sub-stress fusion sequence numbers. The fusion feature sequence of the trend misalignment tensor and the thermal stability focusing factor is a dynamic feature stream composed of multiple fusion feature points. Each feature point carries trend parameters from skin conductance response signals and heart rate variability signals, as well as stability parameters from the temperature signal. The trend extension level describes the time span of the trend misalignment, the trend offset density reflects the frequency of trend changes, and the trend misalignment intensity quantifies the discrete amplitude of the trend. The convergence duration, convergence amplitude contraction rate, and convergence change span collectively reflect the intensity and continuity of the temperature signal's transition from fluctuation to stability. By serializing and concatenating these six-dimensional features in the time dimension, a highly expressive joint change trajectory sequence can be formed. This sequence will be transformed into a sub-stress fusion sequence number in the subsequent coding modeling process, and then used to identify whether an individual is in a sub-stress state.

[0057] S5. When the sub-stress fusion sequence number is identified as a sub-stress state, the sub-stress fusion sequence number is used as the driving variable to match the three-segment meridian treatment trajectory map, and the meridian treatment trajectory is dynamically adjusted according to the real-time changes of the trend misalignment tensor map and the thermal stability focusing factor, so as to complete the continuous execution of TCM external treatment control.

[0058] In this embodiment, S5 specifically refers to: When the sub-stress fusion sequence number is identified as a sub-stress state, the encoded sequence in the sub-stress fusion sequence number is matched with the trajectory segment number in the three-segment meridian treatment trajectory diagram according to the time sequence. By comparing the sequence rhythm of the sub-stress fusion sequence number with the segment rhythm of the three-segment meridian treatment trajectory diagram, the three-segment meridian treatment trajectory diagram that matches the sub-stress fusion sequence number is determined, so that the segment arrangement of the meridian treatment trajectory diagram is matched by the sub-stress fusion sequence number as the driving variable. When a sub-stress fusion sequence number is identified as a sub-stress state, the dynamic adaptation of TCM external treatment strategies needs to be driven by its internal coding sequence. Specifically, the sub-stress fusion sequence number is interpreted as a series of time-sequentially arranged feature codes, each code representing a joint characteristic state of trend deviation and temperature convergence. A pre-defined model of a three-segment meridian treatment trajectory is constructed, dividing the model into three consecutive treatment segments, each with an independent number and corresponding intervention structure. The system matches each set of codes from the sub-stress fusion sequence number with the segment numbers of the three-segment meridian treatment trajectory in chronological order, comparing the rhythmic characteristics of the two in terms of frequency of change, time period, and rhythm density, and selecting the trajectory template with the highest rhythmic consistency. This template serves as the subsequent treatment path, making the matching of the three-segment meridian treatment trajectory no longer static but dynamically responding to the individual's current sub-stress characteristic state.

[0059] The three-segment meridian treatment trajectory map is a set of TCM meridian intervention paths comprising an initiation segment, an enhancement segment, and a consolidation segment. Each segment defines the treatment site, intervention method, and duration parameters, forming a three-dimensional structural sequence. The sequence rhythm of the sub-stress fusion sequence number refers to the distribution rhythm of its internal coding in the time dimension, reflecting the development trajectory of an individual's sub-stress response. The segment rhythm of the three-segment meridian treatment trajectory map represents the treatment design of each intervention segment in terms of time rhythm. By performing feature normalization and similarity comparison of these two rhythms, the path that best matches the current sub-stress fusion sequence number can be identified from multiple preset trajectory maps, ensuring a high degree of consistency between the intervention strategy and the individual's sub-stress evolution stage, thereby improving the timeliness and accuracy of treatment.

[0060] After completing the matching of the three-segment meridian treatment trajectory map, the trend extension level, trend offset density and trend misalignment intensity and the change convergence rate of the thermal stability focusing factor, mean translation rate and amplitude saturation in the trend misalignment tensor map are input into the matching results in time order. By mapping the trend characteristics of the trend misalignment tensor map to the convergence characteristics of the thermal stability focusing factor point by point, a dynamic control factor is formed to adjust the node parameters of the three-segment meridian treatment trajectory map. After matching the three-segment meridian treatment trajectory map, to achieve individualized and precise intervention, the core features of the trend misalignment tensor map and the thermal stability focusing factor need to be synchronously imported into the intervention model to realize real-time control of the node parameters in the three-segment meridian treatment trajectory map. Specifically, the trend extension level, trend offset density, and trend misalignment intensity from the trend misalignment tensor map are first extracted in chronological order and synchronously input into the trajectory map matching results along with the convergence rate, mean shift rate, and amplitude saturation of the thermal stability focusing factor within the corresponding time period. In the data integration stage, a point-by-point mapping mechanism is used to establish a correlation mapping between the six-dimensional feature group formed by the trend features and convergence features at each time point and the node parameters of the three-segment trajectory map. The mapping function calculates the regulatory weight of the current feature on the node stimulation frequency, duration, or intensity, generating a real-time regulation factor. This allows for dynamic adjustment of the treatment trajectory, achieving adaptive updating of the TCM intervention path based on the individual's current physiological signal feedback.

[0061] Trend extension level is an indicator in the trend misalignment tensor plot representing the degree of persistence of angular change differences on the time axis, reflecting the continuity of trend changes; trend offset density measures the distribution density of trend misalignment points within a time period, revealing the degree of offset concentration; trend misalignment intensity characterizes the severity of trend abrupt changes through the combined intensity of angular change differences and directional jump differences. The convergence rate of the thermal stability focusing factor indicates the contraction trend of the temperature signal fluctuation range, the mean shift rate quantifies the overall drift amplitude of the temperature signal average, and the amplitude saturation is used to assess the degree to which temperature peak and valley fluctuations tend to stabilize. By fusing these technical features point by point to form a dynamic control factor, not only can subtle changes in individual physiological states be perceived in real time, but the parameter settings of the three-segment meridian treatment trajectory map at specific nodes can also be precisely adjusted, ensuring that the treatment process is always dynamically matched with the individual's state, improving the accuracy and stability of the intervention effect.

[0062] The dynamic regulation factor is mapped to each trajectory node of the three-segment meridian treatment trajectory map, so that the treatment intensity, treatment rhythm and treatment duration of the trajectory node are updated according to the real-time changes of the trend misalignment tensor map and the thermal stability focusing factor. Under the guidance of the sub-stress fusion sequence number, the meridian treatment trajectory is continuously regulated during the execution process to complete the continuous execution of TCM external treatment control.

[0063] After the three-segment meridian treatment trajectory map is generated, dynamic regulatory factors are mapped to each trajectory node in the trajectory map in real time to achieve adaptive adjustment of intervention parameters. Specifically, based on the time series synchronization of the trend misalignment tensor map and the thermal stability focusing factor, the system assigns the six-dimensional dynamic regulatory factor at each time point to the corresponding trajectory node in the three-segment trajectory map. The system adjusts the treatment intensity based on the trend strength and temperature convergence amplitude in the regulatory factors, controlling the acceleration or deceleration of the treatment rhythm through the combined control of trend extension and convergence duration, while determining the treatment duration based on trend offset density and amplitude saturation. When the sub-stress fusion sequence number identifies an individual still in a sub-stress state, the system continuously calls new trend and temperature data to update the regulatory factors, further updating and iterating the trajectory nodes, ensuring continuity and individual state feedback adaptability throughout the entire TCM external treatment process.

[0064] The intensity of treatment at a trajectory node refers to the magnitude of the physical action applied to an individual's meridians at a specific point in time, typically manifested as the output intensity of heating, massage, or electrical stimulation. Treatment rhythm is the temporal temporal rhythm of the treatment operation, determining the interval and frequency of stimulation output and directly affecting the synchronicity of the individual's nervous system response. Treatment duration represents the duration of the stimulation operation corresponding to each trajectory node, and is a crucial control parameter for achieving the cumulative effect of physiological regulation. By precisely mapping dynamic regulatory factors to these three types of parameters, the treatment pathway can achieve a highly dynamic response. Driven by the sub-stress fusion sequence number, the trajectory map is updated in real time based on individual physiological feedback, enabling the entire external treatment process to have closed-loop control capabilities and improving the timeliness and biocompatibility of treatment intervention.

[0065] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means (e.g., infrared, wireless, microwave, etc.). A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0066] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0067] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0068] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the 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 an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0069] 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.

[0070] In addition, 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.

[0071] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A TCM external treatment control method based on physiological parameter monitoring, characterized in that, Specifically, the following steps are included: S1. Collect skin conductance response signals, heart rate variability signals and temperature signals, and generate a trend-frequency reference surface based on a unified time base, so that the collected skin conductance response signals, heart rate variability signals and temperature signals form a continuous trend trajectory within the trend-frequency reference surface; S2. Extract the turning angle sequence and direction jump sequence of the skin conductance response signal and heart rate variability signal within the trend-frequency reference plane, construct a trend misalignment tensor map, and determine whether there is a trend misalignment between the skin conductance response signal and the heart rate variability signal based on the trend misalignment tensor map. S3. When the trend misalignment tensor plot indicates that a trend misalignment exists, perform continuous window analysis on the temperature signal, calculate the convergence rate of change, the mean shift rate and the amplitude saturation, generate the thermal stability focusing factor, and determine whether the temperature signal exhibits fluctuation stabilization. S4. When the thermal stability focusing factor characterizes the temperature fluctuation to stabilize, the trend characteristics of the fusion trend misalignment tensor plot and the convergence characteristics of the thermal stability focusing factor are used to generate a sub-stress fusion sequence number, and the individual is identified as being in a sub-stress state based on the sub-stress fusion sequence number. S5. When the sub-stress fusion sequence number is identified as a sub-stress state, the sub-stress fusion sequence number is used as the driving variable to match the three-segment meridian treatment trajectory map, and the meridian treatment trajectory is dynamically adjusted according to the real-time changes of the trend misalignment tensor map and the thermal stability focusing factor, so as to complete the continuous execution of TCM external treatment control.

2. The method for controlling external treatment in traditional Chinese medicine based on physiological parameter monitoring according to claim 1, characterized in that, S1 specifically refers to: The device continuously collects three types of physiological parameters—skin conductance response signal, heart rate variability signal, and temperature signal—using a time synchronization acquisition device, and adds a time stamp in a unified format to each type of physiological parameter, so that the skin conductance response signal, heart rate variability signal, and temperature signal have a unified time reference. Under a unified time reference, the skin conductance response signal, heart rate variability signal and temperature signal are aligned according to the time mark, and a trend co-frequency reference surface is constructed in the three-dimensional parameter space with the time reference as a reference, so that the skin conductance response signal, heart rate variability signal and temperature signal form a mappable trend data point set in the trend co-frequency reference surface; Within the trend reference plane, the trend data points of the skin conductance response signal, heart rate variability signal, and temperature signal are continuously interpolated and smoothed to form a continuous trend trajectory in the form of a time series.

3. The method for controlling external treatment in traditional Chinese medicine based on physiological parameter monitoring according to claim 1, characterized in that, S2 specifically includes the following steps: S201. Within the trend reference plane, the continuous trend trajectories of the skin conductance response signal and the heart rate variability signal are divided into time segments, and the trajectory direction change angle is calculated in each time segment to generate the turning angle sequence of the skin conductance response signal and the turning angle sequence of the heart rate variability signal. Based on the change amplitude of the trajectory direction between adjacent time segments, the directional jump sequence of the skin conductance response signal and the directional jump sequence of the heart rate variability signal are formed. S202. Arrange the turning angle sequence of the skin conductance response signal and the turning angle sequence of the heart rate variability signal according to a unified time base, and combine them together with the directional jump sequence of the skin conductance response signal and the directional jump sequence of the heart rate variability signal to form a trend misalignment tensor unit, so that each trend misalignment tensor unit contains the angle change difference and directional jump difference within a time segment, and combine the trend misalignment tensor units in time order to form a trend misalignment tensor graph. S203. By comparing whether the difference in angle change and the difference in direction jump of the continuous trend misalignment tensor units in the trend misalignment tensor plot are continuously greater than the preset difference threshold, when the difference in angle change and the difference in direction jump are greater than the preset difference threshold in multiple continuous trend misalignment tensor units, it is determined that there is a trend misalignment between the skin conductance response signal and the heart rate variability signal.

4. The method for controlling external treatment in traditional Chinese medicine based on physiological parameter monitoring according to claim 3, characterized in that, S203 specifically refers to: Select consecutively arranged trend misalignment tensor units in the trend misalignment tensor plot, and read the angle change difference and direction jump difference of each trend misalignment tensor unit in sequence, so that each trend misalignment tensor unit corresponds to an angle change difference value and a direction jump difference value, so as to form a continuously comparable difference sequence group. The angle change difference and the direction jump difference in the difference sequence group are compared with the preset difference threshold respectively, and the comparison results are marked according to the time order of the trend misalignment tensor unit, so that each trend misalignment tensor unit has a judgment mark for whether the angle change difference is greater than the preset difference threshold and whether the direction jump difference is greater than the preset difference threshold. When all the criteria for the angle change difference and the criteria for the direction jump difference are greater than the preset difference threshold in multiple consecutively arranged trend misalignment tensor units in the difference sequence group, it is determined that there is a trend misalignment between the skin conductance response signal and the heart rate variability signal.

5. The method for controlling external treatment in traditional Chinese medicine based on physiological parameter monitoring according to claim 1, characterized in that, S3 specifically includes the following steps: S301. When the trend misalignment tensor plot indicates that a trend misalignment exists, extract the temperature signal sequence corresponding to the trend misalignment time period, and divide the temperature signal sequence into multiple equal-length continuous time windows, forming a temperature micro-change segment within each time window. S302. Calculate the convergence rate, mean translation rate, and amplitude saturation within each temperature micro-variation segment. The convergence rate is calculated by the reduction ratio of the distance between the maximum and minimum values ​​in the preceding and following temperature micro-variation segments. The mean translation rate is calculated by the average change of the adjacent temperature micro-variation segments. The amplitude saturation is calculated by the degree of amplitude convergence of the continuous temperature peak-valley fluctuations. S303. The convergence rate of change, the mean shift rate and the amplitude saturation are weighted and combined to generate a thermal stability focusing factor to characterize the stable trend of the temperature signal. It is determined whether the value of the thermal stability focusing factor is continuously within the preset value range. If multiple consecutive thermal stability focusing factors are within the preset value range, it is determined that the temperature signal has a fluctuating and stabilizing performance.

6. The method for controlling external treatment in traditional Chinese medicine based on physiological parameter monitoring according to claim 5, characterized in that, S302 specifically refers to: Within each temperature micro-variation segment formed by continuous window analysis, all temperature data points of the segment are traversed, the maximum and minimum temperature values ​​in the segment are recorded, and the distance difference between the maximum and minimum temperature values ​​of adjacent temperature micro-variation segments is compared. This distance difference is used to calculate the change convergence rate, which reflects the degree of contraction in the temperature change amplitude of adjacent temperature micro-variation segments. Between adjacent temperature variation segments, the average temperature in each temperature variation segment is calculated, and the difference between the average temperature of adjacent temperature variation segments is calculated. This difference is used as the basis for calculating the mean translation rate. The mean translation rate reflects the speed of the overall temperature trend and is used to determine whether the temperature change tends to stabilize. Based on the temperature peak and valley data in continuous temperature micro-variation segments, the amplitude range of multiple continuous temperature peak and valley fluctuations is extracted, and the convergence degree between the continuous amplitude ranges is numerically quantified to form amplitude saturation. Amplitude saturation reflects whether the temperature fluctuation shows a stabilization trend, and together with the change convergence rate and mean shift rate, it is used to generate a comprehensive evaluation basis for temperature stability.

7. The method for controlling external treatment in traditional Chinese medicine based on physiological parameter monitoring according to claim 1, characterized in that, S4 specifically includes the following steps: S401. When the thermal stability focusing factor characterizes the temperature fluctuation to stabilize, extract the angle change difference sequence and the direction jump difference sequence of the trend misalignment time period in the trend misalignment tensor. By numerically normalizing the angle change difference sequence and the direction jump difference sequence, obtain the trend extension level, trend offset density and trend misalignment intensity, which are used as trend features of the trend misalignment tensor. S402. Extract the continuous change sequence of the thermal stability focusing factor within the trend misalignment period. Calculate the convergence duration, convergence amplitude contraction rate, and convergence change span of the thermal stability focusing factor to form a convergence feature that characterizes the convergence change process of the temperature signal. Then, fuse this feature with the trend features of the trend misalignment tensor plot point by point according to the time sequence within the trend misalignment period to generate a fusion feature sequence of the trend misalignment tensor plot and the thermal stability focusing factor. S403. Encode each set of trend features and convergence features in the fusion feature sequence to form a fusion coding sequence that reflects the joint pattern of trend shift and temperature convergence. Arrange the fusion coding sequence in chronological order to form a sub-stress fusion sequence number. By comparing the coding arrangement structure of the sub-stress fusion sequence number with the preset sub-stress pattern sequence segment by segment, when the sub-stress fusion sequence number is consistent with the preset sub-stress pattern sequence in a continuous time period, identify the individual as being in a sub-stress state based on the sub-stress fusion sequence number.

8. The method for controlling external treatment in traditional Chinese medicine based on physiological parameter monitoring according to claim 7, characterized in that, S402 specifically refers to: During the trend misalignment period, the continuous change values ​​of the thermal stability focusing factor are traversed in chronological order. The thermal stability focusing factor segments that are continuously and uninterruptedly in the stable range are marked. The convergence duration is calculated by statistically analyzing the start and end times of the segments. The convergence amplitude compression rate is calculated based on the maximum compression amplitude of the numerical fluctuations between adjacent thermal stability focusing factor segments. The convergence change span is calculated based on the maximum change range of the thermal stability focusing factor values ​​during the entire trend misalignment period. The convergence duration, convergence amplitude contraction rate, and convergence change span corresponding to each thermally stable focusing factor segment are combined into a convergence feature group corresponding to the time point. Then, according to each time point in the trend misalignment period, the convergence feature group is fused point by point with the trend extension level, trend offset density, and trend misalignment intensity at the same time point in the trend misalignment tensor to generate the fusion feature point of the trend misalignment tensor and the thermally stable focusing factor. The fusion feature points are sequentially spliced ​​together in chronological order to construct a fusion feature sequence of trend misalignment tensor and thermal stability focusing factor, which serves as the basis for generating subsequent sub-stress fusion sequence numbers.

9. The method for controlling external treatment in traditional Chinese medicine based on physiological parameter monitoring according to claim 1, characterized in that, S5 specifically refers to: When the sub-stress fusion sequence number is identified as a sub-stress state, the encoded sequence in the sub-stress fusion sequence number is matched with the trajectory segment number in the three-segment meridian treatment trajectory diagram according to the time sequence. By comparing the sequence rhythm of the sub-stress fusion sequence number with the segment rhythm of the three-segment meridian treatment trajectory diagram, the three-segment meridian treatment trajectory diagram that matches the sub-stress fusion sequence number is determined, so that the segment arrangement of the meridian treatment trajectory diagram is matched by the sub-stress fusion sequence number as the driving variable. After completing the matching of the three-segment meridian treatment trajectory map, the trend extension level, trend offset density and trend misalignment intensity and the change convergence rate of the thermal stability focusing factor, mean translation rate and amplitude saturation in the trend misalignment tensor map are input into the matching results in time order. By mapping the trend characteristics of the trend misalignment tensor map to the convergence characteristics of the thermal stability focusing factor point by point, a dynamic control factor is formed to adjust the node parameters of the three-segment meridian treatment trajectory map. The dynamic regulation factor is mapped to each trajectory node of the three-segment meridian treatment trajectory map, so that the treatment intensity, treatment rhythm and treatment duration of the trajectory node are updated according to the real-time changes of the trend misalignment tensor map and the thermal stability focusing factor. Under the guidance of the sub-stress fusion sequence number, the meridian treatment trajectory is continuously regulated during the execution process to complete the continuous execution of TCM external treatment control.

Images