Laparoscopic postoperative rehabilitation data acquisition method and system based on traditional Chinese medicine home platform

Abstract

The invention relates to the technical field of data acquisition, and discloses a laparoscopic postoperative rehabilitation data acquisition method and system based on a traditional Chinese medicine home platform. The method comprises the following steps: collecting multi-source rehabilitation data and carrying out traditional Chinese medicine syndrome differentiation analysis to obtain a rehabilitation index; inputting the rehabilitation index into an ARIMA model to predict a rehabilitation trend; discretizing the rehabilitation index to obtain a symbol, constructing a patient specificity matrix, calculating a Hamming distance to identify a rehabilitation mode, calculating an early warning index in combination with doctor intervention time, and generating a personalized rehabilitation risk assessment report. The technical problems that in existing laparoscopic postoperative rehabilitation monitoring, monitoring parameters are single, traditional Chinese medicine theoretical guidance is lacked, and rehabilitation schemes are insufficient in individuation are solved. Through multi-source rehabilitation data acquisition, traditional Chinese medicine syndrome differentiation analysis, ARIMA prediction model and patient specificity matrix construction, comprehensive monitoring, scientific prediction and personalized guidance of the rehabilitation condition of a patient are realized.

Description

Technical Field

[0001] This application relates to the field of data acquisition technology, and in particular to a method and system for acquiring postoperative rehabilitation data of laparoscopic surgery based on a traditional Chinese medicine home care platform. Background Technology

[0002] With the development of minimally invasive techniques, laparoscopic surgery has been widely used in clinical practice due to its advantages such as less trauma and faster recovery. However, monitoring patients' health status and providing rehabilitation guidance after leaving the hospital during the postoperative recovery process still faces challenges. Traditional postoperative follow-up methods mainly rely on patients' regular hospital visits, which cannot achieve real-time monitoring and timely intervention. Existing remote health monitoring systems have begun to be applied to postoperative management, but most focus on monitoring basic physiological parameters and providing simple rehabilitation guidance, lacking the integration and application of traditional Chinese medicine (TCM) theories and diagnostic elements. TCM has unique advantages in postoperative rehabilitation; its holistic concept and principles of syndrome differentiation and treatment can provide patients with individualized rehabilitation plans, but current technologies have not yet fully utilized TCM theories to guide postoperative rehabilitation monitoring after laparoscopy.

[0003] The main shortcomings of existing technologies include: first, the monitoring parameters are limited, mainly focusing on basic physiological parameters such as blood pressure and heart rate, lacking the collection and analysis of TCM diagnostic elements such as tongue appearance, pulse, and complexion; second, data transmission and processing do not fully consider the special characteristics of medical data, lacking differentiated transmission strategies for data of different importance; third, rehabilitation assessment lacks guidance from TCM theory, making it difficult to achieve comprehensive assessment combining TCM and Western medicine; and fourth, rehabilitation guidance lacks personalized and dynamic adjustment mechanisms, making it impossible to adjust rehabilitation plans in a timely manner according to the patient's recovery progress. These problems lead to poor rehabilitation monitoring results after laparoscopic surgery, affecting the quality and efficiency of patient rehabilitation. Summary of the Invention

[0004] This application provides a method and system for collecting laparoscopic postoperative rehabilitation data based on a traditional Chinese medicine home-based platform, which is used to solve the technical problems of single monitoring parameters, lack of guidance from traditional Chinese medicine theory, and insufficient personalization of rehabilitation plans in existing laparoscopic postoperative rehabilitation monitoring.

[0005] Firstly, this application provides a method for collecting postoperative rehabilitation data after laparoscopic surgery based on a TCM home-based platform. The method includes: collecting multi-source rehabilitation data; performing TCM syndrome differentiation analysis on the multi-source rehabilitation data to obtain a TCM rehabilitation index; inputting the TCM rehabilitation index into an ARIMA time series prediction model to predict the rehabilitation trend and obtain a rehabilitation prediction result; discretizing the TCM rehabilitation index to obtain a dispersion degree symbol; constructing an N-row, W-column patient-specific matrix based on the dispersion degree symbol; identifying normal and abnormal rehabilitation patterns by calculating the Hamming distance between each matrix element and the normal value; and calculating a warning index based on the identified abnormal patterns and the inverse relationship between the doctor's intervention time, thereby generating a personalized rehabilitation risk assessment report.

[0006] Secondly, this application provides a laparoscopic postoperative rehabilitation data acquisition system based on a traditional Chinese medicine home care platform, the laparoscopic postoperative rehabilitation data acquisition system based on a traditional Chinese medicine home care platform comprising: The extraction module is used to collect multi-source rehabilitation data, perform TCM syndrome differentiation analysis on the multi-source rehabilitation data, and obtain the TCM rehabilitation index. The prediction module is used to input the TCM rehabilitation index into the ARIMA time series prediction model to predict the rehabilitation trend and obtain the rehabilitation prediction result. The generation module is used to discretize the TCM rehabilitation index to obtain a discreteness symbol. Based on the discreteness, an N-row W-column patient-specific matrix is ​​constructed. By calculating the Hamming distance between each matrix element and the normal value, normal rehabilitation patterns and abnormal rehabilitation patterns are identified. Based on the identified abnormal patterns and the inverse relationship between the doctor's intervention time, a warning index is calculated to generate a personalized rehabilitation risk assessment report.

[0007] Thirdly, a laparoscopic postoperative rehabilitation data acquisition device based on a traditional Chinese medicine home care platform is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory to cause the laparoscopic postoperative rehabilitation data acquisition device based on the traditional Chinese medicine home care platform to execute the above-described laparoscopic postoperative rehabilitation data acquisition method based on the traditional Chinese medicine home care platform.

[0008] Fourthly, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium, which, when executed on a computer, cause the computer to perform the above-described method for collecting laparoscopic postoperative rehabilitation data based on a traditional Chinese medicine home care platform.

[0009] The technical solution provided in this application combines multi-source rehabilitation data collection with TCM syndrome differentiation analysis, effectively integrating Western medicine objective indicators with TCM subjective symptoms. This overcomes the limitations of single parameters in traditional monitoring, providing a comprehensive data foundation and TCM theoretical guidance for postoperative rehabilitation, enabling the TCM rehabilitation index to objectively reflect the patient's overall rehabilitation status. The application of the ARIMA time series prediction model provides a scientific basis for rehabilitation trend prediction. This model fully considers the time series characteristics of rehabilitation data, capturing trends and periodic changes in the data through autoregressive and moving average components, resulting in high accuracy and reliability of rehabilitation prediction results, providing predictive guidance for timely intervention by doctors. The discretization of the rehabilitation index and the introduction of the degree of discretization symbol simplify complex continuous numerical analysis, making rehabilitation data easier to interpret and process, while retaining key characteristic information of the original data. The construction of a patient-specific matrix provides a structured data representation for rehabilitation pattern recognition. The N-row, W-column matrix design considers both the multidimensionality of monitoring parameters and the evolution of the time dimension, making rehabilitation assessment more three-dimensional and dynamic. The method of identifying normal and abnormal rehabilitation patterns by calculating the Hamming distance between matrix elements and normal values ​​effectively distinguishes different rehabilitation paths, accurately captures abnormal states during the rehabilitation process, and improves the accuracy of abnormality detection. The strategy of calculating a warning index based on the inverse relationship between the identified abnormal patterns and physician intervention time enables the quantification and grading of risk assessment, leading to more rational allocation of medical resources and more timely intervention measures. The generation of personalized rehabilitation risk assessment reports integrates the above-mentioned technological features, providing patients and doctors with intuitive and comprehensive assessments of their rehabilitation status and precise rehabilitation guidance suggestions.

[0010] Meanwhile, the time series processing capabilities of the ARIMA prediction model and the pattern recognition technology of the patient-specific matrix have been optimized for specific application scenarios of laparoscopic postoperative rehabilitation. It fully considers the timeliness, volatility and individual differences of rehabilitation data, and makes key contributions to the prediction of TCM rehabilitation index and the identification of abnormal patterns through algorithm features. It significantly improves the accuracy and personalization of rehabilitation assessment, and provides scientific, efficient and personalized rehabilitation monitoring and guidance programs for laparoscopic postoperative patients. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1This is a schematic diagram of an embodiment of the laparoscopic postoperative rehabilitation data collection method based on a traditional Chinese medicine home care platform in this application. Figure 2 This is a schematic diagram of an embodiment of the laparoscopic postoperative rehabilitation data acquisition system based on a traditional Chinese medicine home care platform in this application. Figure 3 This is a schematic block diagram of the structure of the laparoscopic postoperative rehabilitation data acquisition device based on a traditional Chinese medicine home care platform in this embodiment of the invention. Detailed Implementation

[0013] This application provides a method and system for collecting postoperative rehabilitation data after laparoscopy based on a traditional Chinese medicine home care platform. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0014] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the data collection method for postoperative rehabilitation after laparoscopy based on a traditional Chinese medicine home care platform in this application includes: Step S101: Collect multi-source rehabilitation data, perform TCM syndrome differentiation analysis on the multi-source rehabilitation data, and obtain the TCM rehabilitation index; Step S102: Input the TCM rehabilitation index into the ARIMA time series prediction model to predict the rehabilitation trend and obtain the rehabilitation prediction results; Step S103: Discretize the TCM rehabilitation index to obtain the degree of dispersion symbol. Construct an N-row, W-column patient-specific matrix based on the degree of dispersion. Identify normal and abnormal rehabilitation patterns by calculating the Hamming distance between each matrix element and the normal value. Calculate the early warning index based on the identified abnormal patterns and the inverse relationship between the doctor's intervention time, and generate a personalized rehabilitation risk assessment report.

[0015] It is understood that the implementing entity of this application can be a laparoscopic postoperative rehabilitation data acquisition system based on a traditional Chinese medicine home care platform, or it can be a terminal or a server; the specific implementation is not limited here. This application's embodiment uses a server as an example for illustration.

[0016] Specifically, rehabilitation data is collected through a multi-source data acquisition module. This module includes a physiological parameter sensor unit to collect physiological indicators such as blood pressure and blood oxygen, a TCM syndrome acquisition unit to acquire tongue appearance, facial color, and voice features, an environmental parameter acquisition unit to record environmental data, and a user interaction unit to collect subjective symptoms. After median filtering for noise reduction and wavelet transform smoothing, TCM syndrome features are extracted from the raw data. Tongue color and coating features are identified from the tongue diagnosis image, facial color type is extracted from facial color information, and voice characteristics are analyzed from the voice. These features are correlated with physiological parameters to establish mapping rules, forming a multi-dimensional TCM syndrome assessment result. Weights are assigned to different indicators based on the type of surgery and TCM syndrome differentiation, and a TCM rehabilitation index is generated through weighted calculation. The TCM rehabilitation index is input into the ARIMA model to predict the rehabilitation trend. First, the time series is tested for stationarity, and non-stationary series are differencing. Then, the autocorrelation function and partial autocorrelation function are calculated to determine the number of autoregressive terms p, the difference frequency d, and the number of moving average terms q, forming the initial parameters of ARIMA(p,d,q). For example, the time series of the rehabilitation index of a patient after cholecystectomy tends to be stationary after first-order differencing. The autocorrelation function is truncated after lag 1, and the partial autocorrelation function is truncated after lag 2. ARIMA(2,1,1) is selected as a candidate model. Multiple candidate models are constructed, and the optimal model is selected by maximum likelihood estimation and the Akaike information criterion. The randomness of the residual sequence is verified, and the rehabilitation index is predicted for the next 7 to 30 days. The TCM rehabilitation index is discretized. The normal range is defined as Grade A, slightly high as A+, slightly low as A-, moderately high as A++, moderately low as A--, severely high as A+++, and severely low as A---. The threshold of the discrete symbol is adjusted according to historical data to form a syndrome-specific discrete standard. The rehabilitation index is organized into a matrix, converted into a symbol sequence, and segmented into time windows to obtain discrete symbol sequence fragments. An N-row, W-column patient-specific matrix is ​​constructed based on the discrete symbols. The Hamming distance between each pattern and the normal value is calculated to identify normal and abnormal rehabilitation patterns. A warning index is calculated by combining the severity of abnormal patterns with the inverse relationship between the time of doctor intervention, and a personalized rehabilitation risk assessment report is generated.

[0017] In one specific embodiment, the process of performing step S101 may specifically include the following steps: Physiological parameters such as blood pressure, pulse oxygen, body temperature, electrocardiogram, respiration and abdominal circumference are collected through the physiological parameter sensor unit; tongue image, facial color information and voice features are collected through the TCM syndrome collection unit; and environmental data and subjective symptoms are collected through the environmental parameter and user interaction unit to obtain raw multi-source rehabilitation data. The original multi-source rehabilitation data were subjected to median filtering for noise reduction and wavelet transform for smoothing to obtain standardized rehabilitation data. TCM syndrome features are extracted from standardized rehabilitation data, feature recognition is performed on tongue diagnosis images, color analysis is performed on facial color information, and acoustic features are extracted from speech information to obtain digital TCM syndrome features. Establish a mapping rule between digital TCM syndrome characteristics and physiological parameters, establish a correspondence between physiological indicators and TCM syndrome status, and obtain multidimensional TCM syndrome assessment results; Based on the type of laparoscopic surgery and the TCM syndrome differentiation, different weights are assigned to each indicator in the multidimensional TCM syndrome assessment results to obtain a personalized weight configuration scheme for the patient. The TCM rehabilitation index is obtained by weighting the patient's personalized weight configuration scheme with the results of multidimensional TCM syndrome assessment.

[0018] Specifically, the physiological parameter sensor unit collects data on blood pressure, pulse oximetry, body temperature, electrocardiogram, respiration, and abdominal circumference. Blood pressure is measured every 4 hours in the initial postoperative period and 2-3 times daily during the stable period. The pulse oximetry sensor monitors in real-time at a frequency of 1Hz. The body temperature sensor records data 3 times daily. The Traditional Chinese Medicine (TCM) syndrome acquisition unit obtains tongue appearance, facial color, and voice features. The tongue diagnosis acquisition device captures the tongue image in a fasting state, the facial color capture device acquires facial color using a calibrated color chart, and the voice acquisition device records voice features. The environmental parameter acquisition unit records temperature, humidity, and air quality. The user interaction unit allows patients to input subjective symptoms. The collected raw data is processed, using a median filtering algorithm to remove outliers. This algorithm effectively removes sudden noise by sorting the values ​​within the data window and selecting the median value as the output. Subsequently, wavelet transform is applied for smoothing, decomposing the signal into different frequency components, retaining low-frequency information while filtering out high-frequency noise, generating standardized rehabilitation data.

[0019] Traditional Chinese medicine (TCM) syndrome features are extracted from standardized data. For tongue diagnosis images, features such as tongue color, tongue coating color, tongue coating thickness, and tongue shape are extracted; color types are extracted from facial color information; and acoustic features are extracted from speech, analyzing voice strength and clarity. A mapping rule is established between digital TCM syndrome features and physiological parameters, such as the relationship between blood pressure and Qi and blood, and the relationship between heart rate and mental state, forming a multidimensional TCM syndrome assessment result. Weights are assigned according to the type of laparoscopic surgery and TCM syndrome differentiation. For example, after cholecystectomy, patients with hepatobiliary damp-heat syndrome have higher weights for liver function and bile reflux indicators; patients with Qi deficiency syndrome have higher weights for respiratory function and fatigue level indicators, forming a personalized weighting scheme. Finally, the weighting scheme and the TCM syndrome assessment results are weighted and calculated to obtain the TCM rehabilitation index. The TCM rehabilitation index is calculated using a weighted summation method: the scores of each indicator are multiplied by their corresponding weights, summed, divided by the total weights, and finally multiplied by 100 to obtain a rehabilitation index with a maximum score of 100. For example, a patient who underwent cholecystectomy had a blood pressure of 125 / 85 mmHg (score 0.8), oxygen saturation of 98% (score 0.9), and body temperature of 37.1℃ (score 0.7). Their tongue was red with a yellow coating (score 0.6), and their complexion was yellowish (score 0.7). Based on these and other symptoms, the diagnosis was liver and gallbladder damp-heat with qi stagnation. In the weighting scheme, blood pressure had a weight of 0.8, oxygen saturation 0.7, body temperature 0.9, tongue appearance 1.2, and complexion 1.0. The calculation process is as follows: (0.8×0.8)+(0.9×0.7)+(0.7×0.9)+(0.6×1.2)+(0.7×1.0)=2.85, and the total weight is 4.6. Therefore, the TCM rehabilitation index = (2.85÷4.6)×100=62. After adjusting according to the patient's subjective symptom score, the final rehabilitation index is 72, indicating that the rehabilitation status is moderately above average, but the liver and gallbladder damp-heat syndrome has not been completely eliminated.

[0020] In one specific embodiment, the process of performing step S102 may specifically include the following steps: The stationarity test is performed on the time series data of the TCM rehabilitation index to determine whether there is a unit root. The non-stationary series is processed by d-order differencing to obtain the rehabilitation index time series. The autocorrelation function and partial autocorrelation function of the rehabilitation index time series are calculated. By observing the truncation and tailing characteristics of the autocorrelation function plot and the partial autocorrelation function plot, the number of autoregressive terms p and the number of moving average terms q are determined, and the initial parameters of the ARIMA(p,d,q) model are obtained. Multiple candidate models are constructed based on the initial parameters of the ARIMA(p,d,q) model. The maximum likelihood estimation method is applied to each candidate model to calculate the model parameters, and the Akaike Information Criterion is used to evaluate the model fit. The model with the smallest AIC value is selected as the optimal ARIMA model. The residual sequence of the optimal ARIMA model is subjected to white noise test, and the randomness of the residual sequence is verified by Box-Pierce test and Ljung-Box test. The parameters of the models that fail the test are adjusted to obtain the target ARIMA model. Historical data of the rehabilitation index are input into the target ARIMA model, and the rehabilitation index for the next 7 to 30 days is predicted through recursive prediction to obtain a sequence of predicted rehabilitation index values. The root mean square error of the predicted rehabilitation index value sequence and the actual observed value is calculated. The model parameters are dynamically updated based on the error analysis results to obtain the rehabilitation prediction results.

[0021] Specifically, the data collection method for laparoscopic postoperative rehabilitation based on a TCM home-based platform uses the ARIMA time series prediction model to predict rehabilitation trends, solving the problem that traditional postoperative follow-up methods cannot achieve real-time monitoring and timely intervention. The time series data of the TCM rehabilitation index are subjected to a stationarity test. A stationarity test examines whether the statistical characteristics of the time series (such as mean, variance, and autocorrelation) change over time, using the augmented Dickey-Fuller (ADF) test to determine the presence of a unit root. A unit root is the eigenvalue of the time series located on the unit circle; its existence indicates that the series is non-stationary. If the p-value of the ADF test is greater than the significance level (usually 0.05), it indicates the presence of a unit root, requiring differencing; otherwise, it indicates stationarity, allowing direct modeling. The non-stationary series undergoes d-order differencing, which calculates the difference between adjacent observations in the series to eliminate trends. First-order differencing represents the difference between two adjacent observations, second-order differencing represents further differencing of the first-order differencing sequence, and so on. The difference frequency d was determined by repeatedly performing the ADF test until the sequence passed the stationarity test. After differencing, a stationary rehabilitation index time series was obtained, providing a foundation for subsequent modeling.

[0022] The autocorrelation function (ACF) and partial autocorrelation function (PACF) were calculated for the stationary rehabilitation index time series. The ACF describes the correlation between the time series and its lagged versions, while the PACF describes the correlation after removing the influence of intermediate lags. Model parameters were determined by observing the truncation and tailing characteristics of the ACF and PACF plots. If the ACF is truncate while the PACF is tailing, the series is suitable for the MA(q) model, where q is the truncation order of the ACF. If the ACF is tailing while the PACF is truncate, the series is suitable for the AR(p) model, where p is the truncation order of the PACF. If both the ACF and PACF are tailing, the series is suitable for the ARMA(p,q) model. The initial parameters of the ARIMA(p,d,q) model were determined by combining the differencing order d.

[0023] Multiple candidate models are constructed based on the initial parameters, typically choosing combinations where p and q are in the range of 0-2, such as ARIMA(0,1,1), ARIMA(1,1,0), ARIMA(1,1,1), and ARIMA(2,1,1). Maximum likelihood estimation is applied to each candidate model to calculate the model parameters, which finds the parameter values ​​that maximize the probability of the observed data. The Akaike Information Criterion (AIC) is then used to evaluate the model fit. AIC considers both the model's fitting accuracy and the number of parameters, and is calculated as AIC = -2ln(L) + 2k, where L is the maximum likelihood value and k is the number of parameters. A smaller AIC value indicates a better model; therefore, the model with the smallest AIC value is selected as the optimal ARIMA model. The residual sequence of the optimal ARIMA model is then subjected to a white noise test to determine whether the residuals are randomly distributed white noise. The Box-Pierce test and the Ljung-Box test are primarily used; both tests determine whether the residual sequence is white noise by calculating the sum of squares of the residual autocorrelation coefficients. If the p-value is greater than the significance level (usually 0.05), it indicates that the residuals are white noise and the model passes the test; otherwise, the model parameters need to be adjusted and refitted. For models that fail the test, adjustments are made by increasing or decreasing the p- and q-values, or by trying seasonal ARIMA models, until the white noise test is passed and the target ARIMA model is obtained. Historical rehabilitation index data are input into the target ARIMA model, and the rehabilitation index for the next 7 to 30 days is predicted using a recursive prediction method. Recursive prediction is a step-by-step prediction method, first predicting the first future value, then using it as a known value to continue predicting the next value, and so on. This method is suitable for short- to medium-term predictions, especially for relatively gradual changes such as rehabilitation after laparoscopic surgery. The root mean square error (RMSE) is calculated between the predicted rehabilitation index value sequence and the actual observed values. RMSE is the square root of the mean of the squares of the differences between the predicted and actual values, used to assess the prediction accuracy. The model parameters are dynamically updated based on the error analysis results. If the RMSE exceeds the preset threshold, data is collected again and the model is rebuilt to ensure the accuracy of the prediction results.

[0024] In one specific embodiment, the process of executing step S103 may specifically include the following steps: Statistical analysis was conducted on the TCM rehabilitation index to determine the normal range and fluctuation characteristics. Based on clinical data, a normal threshold was set, and the rehabilitation index within the normal range was defined as Grade A, thus obtaining the basic dispersion standard. The degree of the rehabilitation index that deviates from the normal range of the baseline dispersion standard is classified as follows: slightly high index is defined as A+, slightly low index is defined as A-, moderately high index is defined as A++, moderately low index is defined as A--, severely high index is defined as A+++, and severely low index is defined as A---. Based on the rehabilitation prediction results and the patient's historical rehabilitation data, the dispersion degree sign threshold was adjusted in a personalized manner. Differentiated grading standards were set for patients with different TCM syndrome types. The threshold of blood stasis-related indicators was adjusted for patients with Qi deficiency and blood stasis type, and the threshold of damp heat-related indicators was adjusted for patients with liver and gallbladder damp heat type, thus obtaining the syndrome type-specific dispersion standard. The rehabilitation index is organized into a time series data matrix according to the time sequence and the type of monitoring parameters. Each numerical point in the time series data matrix is ​​symbolized by applying the syndrome-specific discrete standard to obtain the discretized symbol sequence of the rehabilitation index. The discretized symbol sequence of the rehabilitation index is segmented by time window. The observation time window W is set, and the discrete symbols at W consecutive time points are combined into sequence segments to obtain multiple discrete symbol sequence segments. Similarity clustering analysis was performed on discrete symbol sequence fragments to identify recurring symbol combination patterns. Patterns that appeared more than twice were assigned clinical significance labels to obtain the discreteness of the symbols.

[0025] Specifically, a statistical analysis of the TCM rehabilitation index was conducted. By calculating the mean and standard deviation of a large amount of historical patient data, the normal range was determined. The normal range is typically defined as the interval between the mean and 0.5 times the standard deviation. Normal thresholds were set based on the type of laparoscopic surgery and the postoperative rehabilitation stage, combined with clinical data. For example, for patients after laparoscopic cholecystectomy, the normal range for the rehabilitation index is 65-75 points in the first week, 75-85 points in the second week, and 85-95 points in the third week. A rehabilitation index within these normal ranges was defined as Grade A, indicating that the patient's recovery progress was in line with expectations, thus obtaining the baseline dispersion standard.

[0026] The rehabilitation index deviating from the normal range of the baseline dispersion standard is classified into multiple levels of dispersion standards. A slightly elevated index (0.5-1 standard deviation above the upper limit of normal) is defined as A+, indicating slightly better-than-expected rehabilitation progress; a slightly low index (0.5-1 standard deviation below the lower limit of normal) is defined as A-, indicating slightly lagging rehabilitation progress; a moderately elevated index (1-1.5 standard deviation above the upper limit of normal) is defined as A++, indicating significantly better-than-expected rehabilitation progress; a moderately low index (1-1.5 standard deviation below the lower limit of normal) is defined as A--, indicating significantly lagging rehabilitation progress; a severely elevated index (more than 1.5 standard deviation above the upper limit of normal) is defined as A+++, indicating far better-than-expected rehabilitation progress; and a severely low index (more than 1.5 standard deviation below the lower limit of normal) is defined as A---, indicating severely lagging rehabilitation progress requiring medical intervention. This multi-level dispersion standard allows for more refined assessment of rehabilitation status. Based on rehabilitation prediction results and patients' historical rehabilitation data, the threshold for the degree of dispersion is adjusted individually to realize the personalized concept of TCM syndrome differentiation and treatment. Differentiated grading standards are set for patients with different TCM syndromes. For example, for patients with Qi deficiency and blood stasis syndrome, the normal range for blood deficiency indicators (such as hemoglobin, complexion, and pulse) is set more leniently, while the normal range for Qi deficiency indicators (such as fatigue, respiration, and voice) is set more strictly. For patients with damp-heat syndrome in the liver and gallbladder, the threshold for damp-heat related indicators (such as tongue coating, gastrointestinal symptoms, and urine color) is set more strictly, while also considering the time process of damp-heat resolution. This personalized adjustment forms a syndrome-specific dispersion standard, which is more in line with the principle of individualized diagnosis and treatment in TCM.

[0027] The rehabilitation index is organized into a time-series data matrix according to chronological order and monitoring parameter type, constructing a multi-dimensional data structure. Rows in the matrix represent different monitoring parameters (e.g., blood pressure, blood oxygen, body temperature, tongue appearance), and columns represent time points (e.g., postoperative days). Each numerical point in the time-series data matrix is ​​symbolically transformed using the previously established syndrome-specific discrete standard, converting continuous numerical variables into discrete symbolic variables, resulting in a discretized symbolic sequence of the rehabilitation index. The discretized symbolic sequence of the rehabilitation index is then segmented into time windows, with an observation time window W (usually 3-7 days). Discrete symbols from W consecutive time points are combined into sequence segments. For example, setting W=5 days, postoperative days 1-5, 6-10, etc., are analyzed as independent sequence segments. This segmentation facilitates the identification of stage-specific characteristics and trends in the rehabilitation process, yielding multiple discrete symbolic sequence segments. The length W of the time window can be adjusted according to the type of laparoscopic surgery and the rehabilitation stage.

[0028] Similarity clustering analysis was performed on discrete symbol sequence fragments, and the Dynamic Time Warping (DTW) algorithm was used to calculate the similarity between sequences and identify recurring symbol combination patterns. The DTW algorithm can handle comparisons of sequences of different lengths, clustering common rehabilitation patterns and assigning clinically meaningful labels to patterns that appear more than twice, such as "Qi and Blood Deficiency Recovery Pattern" and "Damp-Heat Reduction Recovery Pattern," resulting in discrete symbols with interpretability based on Traditional Chinese Medicine (TCM) theory. These symbols not only describe the rehabilitation state but also contain clinical intervention suggestions guided by TCM theory.

[0029] In one specific embodiment, the process of constructing an N-row, W-column patient-specific matrix according to the degree of dispersion can specifically include the following steps: Based on the discreteness symbol sequence, the data are organized into a matrix according to the monitoring parameter type and time order. The number of rows in the matrix is ​​equal to the total number of monitored parameters, and the number of columns in the matrix is ​​equal to the length of the observation time window, thus obtaining a patient-specific matrix. Initial pattern recognition is performed on the patient-specific matrix. The elements of the normal value matrix are defined as the normal rehabilitation standard, that is, all elements are standard normal patterns at the normal level. A candidate pattern set is constructed, and each candidate pattern is a continuous sequence segment in the matrix, resulting in multiple candidate patterns. For each candidate pattern, calculate the Hamming distance with the elements of the normal value matrix. The Hamming distance is defined as the number of different elements at corresponding positions in two sequences. For non-binary symbols, assign different weights to the distance between symbols according to their severity to obtain the Hamming distance value for each candidate pattern. Candidate patterns are classified based on Hamming distance values. A Hamming distance threshold is set. Candidate patterns with a Hamming distance less than or equal to the Hamming distance threshold are classified as normal recovery patterns, and candidate patterns with a Hamming distance greater than the Hamming distance threshold are classified as abnormal recovery patterns, thus obtaining preliminary pattern classification results. The rehabilitation abnormal patterns in the preliminary pattern classification results are further subdivided, and the abnormal patterns are ranked according to the severity of Hamming distance values. The recurring abnormal pattern sequences are extracted as consensus abnormal patterns, resulting in multiple representative rehabilitation abnormal patterns. Clinical significance analysis was conducted on several representative abnormal rehabilitation patterns. The syndrome characteristics of the abnormal patterns were analyzed in combination with the TCM syndrome differentiation theory. The abnormal patterns were mapped to TCM pathogenesis and syndrome to obtain normal rehabilitation patterns and abnormal rehabilitation patterns.

[0030] Specifically, the patient-specific matrix organizes data by monitoring parameter type (rows) and time series (columns). The number of rows N equals the total number of monitored parameters (blood pressure, blood oxygen, etc.), and the number of columns W equals the observation time window length (usually 3-7 days), making the temporal changes of different parameters more intuitive. The elements of the normal value matrix are defined as all standard normal patterns of Grade A (normal level) as ideal references. Row vectors, column vectors, or submatrices in the matrix are extracted as candidate patterns using a sliding window method.

[0031] For each candidate pattern, the Hamming distance to the normal value matrix is ​​calculated, which is the number of different elements at corresponding positions in the two sequences. Considering that discrete symbols have multiple levels (A, A+, A-, etc.), a weighted Hamming distance is introduced. The weight of A and A+ / A- is 1, the weight of A and A++ / A-- is 2, and the weight of A and A+++ / A--- is 3, thus more accurately reflecting the degree of deviation. Based on the Hamming distance values, a threshold is set (usually 20% of the total number of matrix elements). Patterns with a distance less than or equal to the threshold are considered normal recovery patterns, and those with a distance greater than the threshold are considered abnormal recovery patterns. Abnormal patterns are sorted by distance, and patterns that occur repeatedly in multiple patients are extracted as consensus abnormal patterns. The pathogenesis characteristics of abnormal patterns are analyzed in conjunction with traditional Chinese medicine syndrome differentiation theory, and they are mapped to traditional Chinese medicine syndromes (such as qi deficiency and blood stasis, damp-heat in the liver and gallbladder, etc.), forming a clinically significant rehabilitation pattern determination.

[0032] In one specific embodiment, the process of calculating the early warning index based on the identified abnormal pattern and the inverse relationship between the time to doctor intervention can specifically include the following steps: The severity of the discrete symbols in the abnormal rehabilitation pattern is graded, and different severity weights are assigned according to the symbol level. Mild abnormal symbols are assigned low weight, moderate abnormal symbols are assigned medium weight, and severe abnormal symbols are assigned high weight, resulting in an abnormal symbol weight comparison table. Determine the upper limit of doctor intervention time corresponding to different degrees of abnormality: long intervention time for mild abnormalities, medium intervention time for moderate abnormalities, and short intervention time for severe abnormalities. Establish an inverse mapping relationship between intervention time and severity to obtain an intervention time reference table. Based on the abnormal symbol weight comparison table and the intervention time comparison table, a formula for calculating the early warning index is established. The severity weight of the abnormal symbol is multiplied by the reciprocal of the corresponding intervention time, and different weight coefficients are applied to the types of monitoring parameters to obtain the early warning index calculation model. The abnormal rehabilitation pattern is input into the early warning index calculation model to obtain a quantified early warning index value. The early warning index is then classified into three levels: low warning, medium warning, and high warning, resulting in the early warning level result. By combining the results of the early warning level and the TCM syndrome interpretation of the abnormal rehabilitation pattern, a personalized rehabilitation suggestion library is constructed, including TCM conditioning plan, acupuncture point recommendation, diet therapy plan, functional exercise and emotional regulation suggestions. Suitable rehabilitation guidance plans are selected from the suggestion library according to the syndrome type and early warning level to obtain personalized rehabilitation guidance suggestions. The results of the early warning level, the analysis of abnormal rehabilitation patterns, and the personalized rehabilitation guidance suggestions are integrated into a systematic report, which includes rehabilitation trend charts, early warning indicators of abnormal points, TCM syndrome differentiation analysis results, and precise rehabilitation guidance, resulting in a personalized rehabilitation risk assessment report.

[0033] Specifically, the severity of discrete symbols in the rehabilitation abnormality pattern is graded, and different weight values ​​are assigned according to the symbol level: mild abnormality symbols (A+ / A-) are assigned a weight of 1, moderate abnormality symbols (A++ / A--) are assigned a weight of 2, and severe abnormality symbols (A+++ / A---) are assigned a weight of 3, forming an abnormality symbol weight reference table. Next, the upper limit of physician intervention time corresponding to different degrees of abnormality is determined, reflecting the traditional Chinese medicine principle of "treating the symptoms in urgent cases." Mild abnormalities usually do not require emergency intervention, corresponding to an intervention time limit of 72 hours; moderate abnormalities require timely treatment, corresponding to an intervention time limit of 24 hours; severe abnormalities require emergency intervention, corresponding to an intervention time limit of 6 hours. This setting establishes an inverse mapping relationship between intervention time and severity, forming an intervention time reference table.

[0034] Based on the two comparison tables mentioned above, a warning index calculation model was established: the warning index equals the product of the severity weight of the abnormal symbol and the reciprocal of the intervention time, multiplied by the importance weight coefficient of the monitoring parameter. High weights (1.2-1.5) were assigned to vital signs such as blood pressure and blood oxygen saturation, medium weights (0.8-1.2) to TCM characteristics such as tongue appearance and facial complexion, and low weights (0.5-0.8) to environmental parameters. The rehabilitation abnormality pattern was input into the warning index calculation model to obtain a quantitative warning index value, which was then graded: below 0.3 indicates a low-level warning, requiring regular follow-up; 0.3-0.7 indicates a moderate warning, requiring enhanced monitoring; and above 0.7 indicates a high-level warning, requiring immediate intervention. This grading method forms the warning level results, guiding subsequent rehabilitation interventions.

[0035] A personalized rehabilitation suggestion database is constructed by combining early warning levels and TCM syndrome interpretation, and rehabilitation programs are designed for different syndrome types. For patients with Qi deficiency and blood stasis, herbs such as Astragalus membranaceus and Angelica sinensis are recommended to tonify Qi and promote blood circulation; for patients with damp-heat in the liver and gallbladder, herbs such as Bupleurum chinense and Artemisia capillaris are recommended to clear heat and eliminate dampness. The database also includes acupuncture point recommendations, dietary therapy plans, and functional exercise suggestions, forming a comprehensive rehabilitation guidance system. Early warning levels, abnormal pattern analysis, and rehabilitation guidance suggestions are integrated into a personalized rehabilitation risk assessment report, realizing full-process management from data collection and abnormal detection to personalized intervention, and solving the problem of lack of real-time monitoring and precise guidance in traditional postoperative follow-up.

[0036] In one specific embodiment, the process of constructing a personalized rehabilitation suggestion library may specifically include the following steps: Based on the abnormal rehabilitation patterns, a TCM syndrome differentiation analysis was conducted, and the abnormal patterns were summarized into different syndrome types to obtain a postoperative syndrome classification system. A database of traditional Chinese medicine (TCM) conditioning solutions was constructed based on the postoperative syndrome classification system. Formulas corresponding to different treatment methods were selected for different syndrome types to obtain TCM conditioning solutions. Based on the postoperative syndrome classification system, targeted acupoint combinations are selected, and corresponding acupoints with different effects are selected for different syndromes to obtain acupuncture point recommendations; Based on the postoperative syndrome classification system, corresponding ingredients and cooking methods are matched, and dietary therapy is recommended for different syndromes to obtain dietary therapy plans. Based on the postoperative syndrome classification system, rehabilitation exercise strategies are developed, and exercise forms and intensities are designed for different syndromes to obtain functional exercise programs; The intensity of traditional Chinese medicine treatment plans, acupuncture point recommendations, dietary therapy plans, and functional exercise plans is graded according to the warning level to obtain a personalized rehabilitation suggestion database.

[0037] Specifically, based on the abnormal recovery patterns, a TCM syndrome differentiation analysis was conducted. Pattern recognition technology was used to match the abnormal patterns with TCM symptom characteristics, categorizing them into common syndrome types such as Qi deficiency and blood stasis, liver and gallbladder damp-heat, spleen and stomach weakness, and Qi stagnation and blood stasis, forming a syndrome classification system after laparoscopic surgery. A corresponding TCM treatment plan library was developed for the constructed postoperative syndrome classification system. Based on the principle of "syndrome differentiation and treatment," formulas for tonifying Qi and activating blood circulation, such as Buyang Huanwu Decoction and Bazhen Decoction, were selected for Qi deficiency and blood stasis; formulas for clearing heat and eliminating dampness, such as Yinchenhao Decoction and Longdan Xiegan Decoction, were selected for liver and gallbladder damp-heat; formulas for strengthening the spleen and stomach, such as Liujunzi Decoction and Shenling Baizhu Powder, were selected for spleen and stomach weakness; and formulas for regulating Qi and activating blood circulation, such as Chaihu Shugan Powder and Xuefu Zhuyu Decoction, were selected for Qi stagnation and blood stasis. These formulas can be adjusted according to the patient's specific condition to ensure the precision of TCM treatment.

[0038] Based on the postoperative syndrome classification system, targeted acupoint combinations were selected to supplement traditional Chinese medicine treatment with acupuncture. For the Qi deficiency and blood stasis type, acupoints such as Zusanli, Qihai, and Xuehai were mainly selected to tonify Qi and promote blood circulation; for the liver and gallbladder damp-heat type, acupoints such as Yanglingquan, Taichong, and Xingjian were selected to soothe the liver and promote gallbladder function; for the spleen and stomach deficiency type, acupoints such as Zhongwan, Pishu, and Weishu were selected to strengthen the spleen and stomach; and for the Qi stagnation and blood stasis type, acupoints such as Taichong, Geshu, and Qimen were selected to regulate Qi and promote blood circulation. These acupoint combinations were selected according to the theory of meridians in traditional Chinese medicine and are highly targeted.

[0039] Based on the postoperative syndrome classification system, corresponding dietary therapy plans are matched, taking into account the properties and effects of foods. For the Qi deficiency and blood stasis type, warming and nourishing dietary therapies such as astragalus stewed chicken and angelica, ginger, and mutton soup are recommended; for the liver and gallbladder damp-heat type, heat-clearing and dampness-removing dietary therapies such as mung bean and barley porridge and crucian carp and tofu soup are recommended; for the spleen and stomach weakness type, spleen-strengthening dietary therapies such as yam porridge and lotus seed and lily bulb porridge are recommended; and for the Qi stagnation and blood stasis type, qi-regulating and blood-activating dietary therapies such as rose tea and hawthorn water are recommended. The dietary therapy plans take into account both nutritional and pharmacological effects and are easy to implement at home. Rehabilitation exercise strategies are formulated based on the postoperative syndrome classification system, designing suitable exercise forms and intensities according to different syndrome types. For the Qi deficiency and blood stasis type, gentle aerobic exercises such as slow walking and Tai Chi are designed; for the liver and gallbladder damp-heat type, liver-soothing and depression-relieving exercises such as "Two Hands Supporting the Sky to Regulate the Three Jiaos" from the Eight Pieces of Brocade are designed; for the spleen and stomach weakness type, abdominal massage and light functional exercises are designed; and for the Qi stagnation and blood stasis type, stretching exercises such as "Bear Movement" from the Five Animal Frolics are designed. These functional exercise programs follow a gradual approach and are suitable for different stages of postoperative rehabilitation.

[0040] The traditional Chinese medicine treatment plan, acupuncture point recommendations, diet therapy plan, and functional exercise plan are classified into intensity levels according to the warning level. The mild treatment plan is selected for the low warning level, the moderate intensity treatment plan is selected for the medium warning level, and the intensive treatment plan is selected for the high warning level, and it is recommended to seek medical attention in time.

[0041] The above describes the method for collecting laparoscopic postoperative rehabilitation data based on a traditional Chinese medicine home care platform in the embodiments of this application. The following describes the laparoscopic postoperative rehabilitation data collection system based on a traditional Chinese medicine home care platform in the embodiments of this application. Please refer to [link to relevant documentation]. Figure 2 One embodiment of the laparoscopic postoperative rehabilitation data acquisition system based on a traditional Chinese medicine home care platform in this application includes: Extraction module 201 is used to collect multi-source rehabilitation data, perform TCM syndrome differentiation analysis on the multi-source rehabilitation data, and obtain TCM rehabilitation index; Prediction module 202 is used to input the TCM rehabilitation index into the ARIMA time series prediction model to predict the rehabilitation trend and obtain the rehabilitation prediction result; The generation module 203 is used to discretize the TCM rehabilitation index to obtain a discreteness symbol, construct an N-row W-column patient-specific matrix based on the discreteness, identify normal rehabilitation patterns and abnormal rehabilitation patterns by calculating the Hamming distance between each matrix element and the normal value, and calculate the early warning index based on the identified abnormal patterns and the inverse relationship between the doctor's intervention time, and generate a personalized rehabilitation risk assessment report.

[0042] above Figure 2 The laparoscopic postoperative rehabilitation data acquisition system based on a traditional Chinese medicine home care platform in this embodiment of the invention will be described in detail from the perspective of modular functional entities. The laparoscopic postoperative rehabilitation data acquisition device based on a traditional Chinese medicine home care platform in this embodiment of the invention will be described in detail from the perspective of hardware processing.

[0043] Figure 3 This is a schematic diagram of the structure of a laparoscopic postoperative rehabilitation data acquisition device based on a traditional Chinese medicine home care platform, according to an embodiment of the present invention. This laparoscopic postoperative rehabilitation data acquisition device 300 based on the traditional Chinese medicine home care platform can vary considerably due to different configurations or performance. It may include one or more central processing units (CPUs) 310 (e.g., one or more processors) and a memory 320, and one or more storage media 330 (e.g., one or more mass storage devices) for storing applications 333 or data 332. The memory 320 and storage media 330 can be temporary or persistent storage. The program stored in the storage media 330 may include one or more modules (not shown in the diagram), and each module may include a series of instruction operations on the laparoscopic postoperative rehabilitation data acquisition device 300 based on the traditional Chinese medicine home care platform. Furthermore, the processor 310 can be configured to communicate with the storage medium 330 and execute a series of instructions in the storage medium 330 on the laparoscopic postoperative rehabilitation data acquisition device 300 based on the TCM home platform to realize the steps of the above-mentioned laparoscopic postoperative rehabilitation data acquisition method based on the TCM home platform.

[0044] The laparoscopic postoperative rehabilitation data acquisition device 300 based on a traditional Chinese medicine home-based platform may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input / output interfaces 360, and / or one or more operating systems 331, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 3 The illustrated structure of the laparoscopic postoperative rehabilitation data acquisition device based on a TCM home care platform does not constitute a limitation on the laparoscopic postoperative rehabilitation data acquisition device based on a TCM home care platform provided by this invention. It may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0045] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the method for collecting laparoscopic postoperative rehabilitation data based on a traditional Chinese medicine home platform.

[0046] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0047] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a laparoscopic postoperative rehabilitation data acquisition device (which can be a personal computer, server, or network device, etc.) based on a traditional Chinese medicine home-based platform to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for collecting postoperative rehabilitation data after laparoscopy based on a traditional Chinese medicine home-based platform, characterized in that, The method includes: Collect multi-source rehabilitation data, perform TCM syndrome differentiation analysis on the multi-source rehabilitation data, and obtain the TCM rehabilitation index; The TCM rehabilitation index is input into the ARIMA time series prediction model to predict the rehabilitation trend and obtain the rehabilitation prediction results. The TCM rehabilitation index is discretized to obtain a discreteness symbol. Based on the discreteness, an N-row, W-column patient-specific matrix is ​​constructed. By calculating the Hamming distance between each matrix element and the normal value, normal and abnormal rehabilitation patterns are identified. Based on the identified abnormal patterns and the inverse relationship between the doctor's intervention time, a warning index is calculated to generate a personalized rehabilitation risk assessment report.

2. The method for collecting postoperative rehabilitation data after laparoscopy based on a traditional Chinese medicine home care platform according to claim 1, characterized in that, The process involves collecting multi-source rehabilitation data, performing TCM syndrome differentiation analysis on the multi-source rehabilitation data, and obtaining a TCM rehabilitation index, including: Physiological parameters such as blood pressure, pulse oxygen, body temperature, electrocardiogram, respiration and abdominal circumference are collected through the physiological parameter sensor unit; tongue image, facial color information and voice features are collected through the TCM syndrome collection unit; and environmental data and subjective symptoms are collected through the environmental parameter and user interaction unit to obtain raw multi-source rehabilitation data. The original multi-source rehabilitation data were subjected to median filtering for noise reduction and wavelet transform for smoothing to obtain standardized rehabilitation data. TCM syndrome features are extracted from the standardized rehabilitation data, feature recognition is performed on the tongue diagnosis image, color analysis is performed on the facial color information, and acoustic features are extracted from the speech information to obtain digital TCM syndrome features. Establish a correlation mapping rule between the digital TCM syndrome characteristics and physiological parameters, establish a correspondence between physiological indicators and TCM syndrome status, and obtain multidimensional TCM syndrome assessment results; Based on the type of laparoscopic surgery and the TCM syndrome differentiation, different weights are assigned to each indicator in the multidimensional TCM syndrome assessment results to obtain a personalized weight configuration scheme for the patient. The TCM rehabilitation index is obtained by weighting the patient's personalized weight configuration scheme with the multidimensional TCM syndrome assessment results.

3. The method for collecting postoperative rehabilitation data after laparoscopy based on a traditional Chinese medicine home care platform according to claim 1, characterized in that, The step of inputting the TCM rehabilitation index into the ARIMA time series prediction model to predict the rehabilitation trend and obtain the rehabilitation prediction results includes: The time series data of the TCM rehabilitation index are subjected to a stationarity test to determine whether there is a unit root. The non-stationary series are subjected to d-order differencing to obtain the rehabilitation index time series. The autocorrelation function and partial autocorrelation function are calculated for the time series of the rehabilitation index. By observing the truncation and tailing characteristics of the autocorrelation function plot and the partial autocorrelation function plot, the number of autoregressive terms p and the number of moving average terms q are determined, and the initial parameters of the ARIMA(p,d,q) model are obtained. Multiple candidate models are constructed based on the initial parameters of the ARIMA(p,d,q) model. The maximum likelihood estimation method is applied to each candidate model to calculate the model parameters, and the Akaike Information Criterion is used to evaluate the model fit. The model with the smallest AIC value is selected as the optimal ARIMA model. The residual sequence of the optimal ARIMA model is subjected to white noise test, and the randomness of the residual sequence is verified by Box-Pierce test and Ljung-Box test. The parameters of the models that fail the test are adjusted to obtain the target ARIMA model. Historical data of the rehabilitation index are input into the target ARIMA model, and the rehabilitation index for the next 7 to 30 days is predicted through recursive prediction to obtain a sequence of predicted rehabilitation index values. The root mean square error is calculated between the predicted rehabilitation index value sequence and the actual observed values. The model parameters are dynamically updated based on the error analysis results to obtain the rehabilitation prediction results.

4. The method for collecting postoperative rehabilitation data after laparoscopy based on a traditional Chinese medicine home care platform according to claim 1, characterized in that, The discretization of the TCM rehabilitation index to obtain the degree of discretization symbol includes: Statistical analysis was performed on the TCM rehabilitation index to determine the normal range and fluctuation characteristics. A normal value threshold was set based on clinical data. The rehabilitation index within the normal range was defined as Grade A, thus obtaining the basic dispersion standard. The degree of the rehabilitation index that deviates from the normal range of the basic dispersion standard is classified as follows: slightly high index is defined as A+, slightly low index is defined as A-, moderately high index is defined as A++, moderately low index is defined as A--, severely high index is defined as A+++, and severely low index is defined as A---. Based on the rehabilitation prediction results and the patient's historical rehabilitation data, the dispersion sign threshold is adjusted in a personalized manner. Differentiated grading standards are set for patients with different TCM syndrome types. For patients with Qi deficiency and blood stasis, the threshold of blood stasis-related indicators is adjusted, and for patients with liver and gallbladder damp-heat, the threshold of damp-heat-related indicators is adjusted to obtain syndrome-specific dispersion standards. The rehabilitation index is organized into a time-series data matrix according to the time sequence and monitoring parameter type. The syndrome-specific discrete standard is applied to each numerical point in the time-series data matrix to perform symbolic transformation, resulting in a discretized symbolic sequence of the rehabilitation index. The time window segmentation process is performed on the discretized symbol sequence of the rehabilitation index. An observation time window W is set, and the discrete symbols at consecutive time points W are combined into sequence segments to obtain multiple discrete symbol sequence segments. Similarity clustering analysis is performed on the discrete symbol sequence fragments to identify recurring symbol combination patterns. Patterns that appear more than twice are assigned clinical significance labels to obtain the discreteness symbols.

5. The method for collecting postoperative rehabilitation data after laparoscopy based on a traditional Chinese medicine home care platform according to claim 4, characterized in that, The step of constructing an N-row, W-column patient-specific matrix based on the degree of dispersion, and identifying normal and abnormal rehabilitation patterns by calculating the Hamming distance between each matrix element and the normal value, includes: Based on the discreteness symbol sequence, the data is organized into a matrix according to the monitoring parameter type and time order. The number of rows in the matrix is ​​equal to the total number of monitored parameters, and the number of columns in the matrix is ​​equal to the length of the observation time window, thus obtaining a patient-specific matrix. Initial pattern recognition is performed on the patient-specific matrix. The elements of the normal value matrix are defined as the normal rehabilitation standard, that is, all elements are standard normal patterns at the normal level. A candidate pattern set is constructed, and each candidate pattern is a continuous sequence segment in the matrix, resulting in multiple candidate patterns. For each of the multiple candidate patterns, the Hamming distance is calculated with the elements of the normal value matrix. The Hamming distance is defined as the number of different elements at corresponding positions in two sequences. For non-binary symbols, the distance between symbols is assigned different weights according to the severity to obtain the Hamming distance value for each candidate pattern. Candidate patterns are classified according to the Hamming distance value. A Hamming distance threshold is set. Candidate patterns with a Hamming distance less than or equal to the Hamming distance threshold are classified as normal recovery patterns, and candidate patterns with a Hamming distance greater than the Hamming distance threshold are classified as abnormal recovery patterns, thus obtaining preliminary pattern classification results. The rehabilitation abnormal patterns in the preliminary pattern classification results are further subdivided, and the abnormal patterns are ranked according to the severity of Hamming distance values. The recurring abnormal pattern sequences are extracted as consensus abnormal patterns to obtain multiple representative rehabilitation abnormal patterns. Clinical significance analysis was performed on the aforementioned representative abnormal rehabilitation patterns. The syndrome characteristics of the abnormal patterns were analyzed in combination with the TCM syndrome differentiation theory. The abnormal patterns were mapped to TCM pathogenesis and syndrome to obtain normal rehabilitation patterns and abnormal rehabilitation patterns.

6. The method for collecting laparoscopic postoperative rehabilitation data based on a traditional Chinese medicine home care platform according to claim 5, characterized in that, The process involves calculating a warning index based on the identified abnormal patterns and the inverse relationship between the time to doctor intervention, generating a personalized rehabilitation risk assessment report, including: The severity of the discrete symbols in the abnormal rehabilitation pattern is graded, and different severity weights are assigned according to the symbol level. Mild abnormal symbols are assigned low weights, moderate abnormal symbols are assigned medium weights, and severe abnormal symbols are assigned high weights, resulting in an abnormal symbol weight comparison table. Determine the upper limit of doctor intervention time corresponding to different degrees of abnormality: long intervention time for mild abnormalities, medium intervention time for moderate abnormalities, and short intervention time for severe abnormalities. Establish an inverse mapping relationship between intervention time and severity to obtain an intervention time reference table. Based on the abnormal symbol weight comparison table and the intervention time comparison table, an early warning index calculation formula is established. The severity weight of the abnormal symbol is multiplied by the reciprocal of the corresponding intervention time, and different weight coefficients are applied to the types of monitoring parameters to obtain the early warning index calculation model. The abnormal rehabilitation pattern is input into the early warning index calculation model to obtain a quantified early warning index value. The early warning index is then classified into three levels: low warning, medium warning, and high warning, to obtain the early warning level result. Based on the TCM syndrome interpretation of the warning level results and the abnormal rehabilitation pattern, a personalized rehabilitation suggestion library is constructed, including TCM conditioning plan, acupuncture point recommendation, diet therapy plan, functional exercise and emotional regulation suggestion. Suitable rehabilitation guidance plan is selected from the suggestion library according to syndrome type and warning level to obtain personalized rehabilitation guidance suggestions. The early warning level results, abnormal rehabilitation pattern analysis, and personalized rehabilitation guidance suggestions are integrated into a systematic report, which includes rehabilitation trend charts, early warning indicators for abnormal points, TCM syndrome differentiation analysis results, and precise rehabilitation guidance, resulting in a personalized rehabilitation risk assessment report.

7. The method for collecting laparoscopic postoperative rehabilitation data based on a traditional Chinese medicine home care platform according to claim 6, characterized in that, The method combines the early warning level results with the TCM syndrome interpretation of abnormal rehabilitation patterns to construct a personalized rehabilitation suggestion database, including: Based on the described abnormal rehabilitation patterns, a TCM syndrome differentiation analysis was conducted, and the abnormal patterns were summarized into different syndrome types to obtain a postoperative syndrome classification system. A TCM treatment plan library was constructed based on the postoperative syndrome classification system. Formulas corresponding to different syndromes were selected to obtain TCM treatment plans. Based on the postoperative syndrome classification system, targeted acupoint combinations are selected, and corresponding acupoints with different effects are selected for different syndromes to obtain acupuncture point recommendations. Based on the postoperative syndrome classification system, corresponding ingredients and cooking methods are matched, and dietary therapy is recommended for different syndromes to obtain dietary therapy plans. Based on the postoperative syndrome classification system, rehabilitation exercise strategies are developed, and functional exercise programs are obtained by designing exercise forms and intensities for different syndromes. The traditional Chinese medicine treatment plan, acupuncture point recommendations, dietary therapy plan, and functional exercise plan are classified into intensity levels according to the warning level to obtain a personalized rehabilitation suggestion library.

8. A data acquisition system for postoperative rehabilitation of laparoscopic patients based on a traditional Chinese medicine home-based platform, characterized in that, The method for collecting laparoscopic postoperative rehabilitation data based on a traditional Chinese medicine home care platform as described in any one of claims 1-7, wherein the laparoscopic postoperative rehabilitation data collection system based on a traditional Chinese medicine home care platform comprises: The extraction module is used to collect multi-source rehabilitation data, perform TCM syndrome differentiation analysis on the multi-source rehabilitation data, and obtain the TCM rehabilitation index. The prediction module is used to input the TCM rehabilitation index into the ARIMA time series prediction model to predict the rehabilitation trend and obtain the rehabilitation prediction result. The generation module is used to discretize the TCM rehabilitation index to obtain a discreteness symbol. Based on the discreteness, an N-row W-column patient-specific matrix is ​​constructed. By calculating the Hamming distance between each matrix element and the normal value, normal rehabilitation patterns and abnormal rehabilitation patterns are identified. Based on the identified abnormal patterns and the inverse relationship between the doctor's intervention time, a warning index is calculated to generate a personalized rehabilitation risk assessment report.

9. A data acquisition device for postoperative rehabilitation of laparoscopic patients based on a traditional Chinese medicine home-based platform, characterized in that, The device includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the method for collecting laparoscopic postoperative rehabilitation data based on a traditional Chinese medicine home care platform as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, the processor executes the method for collecting laparoscopic postoperative rehabilitation data based on a traditional Chinese medicine home care platform as described in any one of claims 1 to 7.

Images