Personalized traditional Chinese medicine enema curative effect dynamic optimization system and method and storage medium

Abstract

The invention provides a personalized traditional Chinese medicine enema curative effect optimization dynamic system and method and a storage medium, relates to the technical field of traditional Chinese medicine informatization, and is used for solving at least one of the technical problems of high subjectivity of traditional Chinese medicine enema curative effect evaluation, lack of dynamic quantitative data support and treatment scheme adjustment hysteresis in related technologies. The method comprises the following steps: acquiring periodical symptomatic quantification data, periodical tongue picture image data and traditional Chinese medicine enema treatment scheme data which represent the current health state of a patient; generating color space data; obtaining a tongue color feature value and a tongue coating texture feature value of the tongue body region; generating a comprehensive curative effect index representing the comprehensive curative effect of the cycle; constructing a time sequence of the comprehensive curative effect indexes; determining a curative effect innovation coefficient and a curative effect imitation coefficient of the model; generating predicted efficacy trends for one or more cycles in the future; and generating a treatment scheme adjustment signal for indicating adjustment of the traditional Chinese medicine enema treatment scheme.

Description

Technical Field

[0001] This application relates to the field of information technology in traditional Chinese medicine, and in particular to a personalized system, method and storage medium for dynamic optimization of the efficacy of traditional Chinese medicine enema. Background Technology

[0002] Traditional Chinese medicine retention enema therapy is an important method of treating internal diseases externally in traditional Chinese medicine, showing unique advantages and good clinical efficacy, especially in the treatment of chronic intestinal diseases such as ulcerative colitis and Crohn's disease. This therapy administers medication through the rectum, allowing the drug to act directly on the lesion site, and has advantages such as rapid absorption, high bioavailability, and avoidance of the first-pass effect of the liver.

[0003] In current clinical practice, the treatment plan for traditional Chinese medicine enemas is usually formulated by a TCM practitioner based on the results of the "four diagnostic methods" (inspection, auscultation and olfaction, inquiry, and palpation) combined with their personal clinical experience. Once the treatment plan is determined, it often remains unchanged for a relatively long treatment cycle (such as one week or several weeks). The physician's evaluation of the efficacy mainly relies on regular follow-up visits, through inquiring about changes in the patient's symptoms and observing the tongue and pulse.

[0004] However, this traditional treatment model has some inherent limitations. First, the continuity and dynamism of efficacy assessment are insufficient. Due to the long intervals between follow-up visits, physicians find it difficult to monitor subtle fluctuations in the patient's condition during the treatment cycle, potentially missing the optimal time to adjust the treatment plan. Second, the assessment process is highly subjective, relying heavily on the physician's personal experience and lacking unified, quantifiable, objective standards, which limits the accuracy and repeatability of efficacy assessment. Third, adjustments to the treatment plan are often reactive, meaning they are only made after poor efficacy or adverse reactions are observed, lacking proactive prediction and optimization. Summary of the Invention

[0005] This application provides a personalized Chinese medicine enema efficacy dynamic optimization system, method, and storage medium to improve at least one of the technical problems in related technologies, namely, the strong subjectivity of Chinese medicine enema efficacy evaluation, the lack of dynamic quantitative data support, and the lag in treatment plan adjustment.

[0006] To achieve the above objectives, the embodiments of this application adopt the following technical solutions:

[0007] In a first aspect, this application provides a method for dynamically optimizing the efficacy of personalized traditional Chinese medicine enema, comprising: acquiring periodic symptomatic quantitative data and periodic tongue image data characterizing the patient's current health status, as well as traditional Chinese medicine enema treatment plan data corresponding to the period; for each periodic tongue image data, extracting the tongue body region from the tongue image, and converting the image data of the tongue body region to a preset color space to generate color space data; and based on the color space data, acquiring the tongue color feature value and tongue coating texture feature value of the tongue body region respectively;

[0008] For each cycle, the periodic symptom-based quantitative data, the tongue color feature value, and the tongue coating texture feature value are pre-combined to generate a comprehensive efficacy index characterizing the overall efficacy of that cycle. Based on the comprehensive efficacy index for multiple cycles, a time series of the comprehensive efficacy index is constructed. The time series of the comprehensive efficacy index is fitted to a pre-defined efficacy cumulative diffusion model to determine the efficacy innovation coefficient and efficacy imitation coefficient of the model. The efficacy innovation coefficient and efficacy imitation coefficient are substituted into the efficacy cumulative diffusion model to generate a predicted efficacy trend for one or more future cycles. The predicted efficacy trend is compared with a pre-defined ideal recovery curve to generate a treatment plan adjustment signal for indicating adjustments to the traditional Chinese medicine enema treatment plan.

[0009] In one possible implementation of the first aspect, the step of obtaining the tongue color feature value and the tongue coating texture feature value of the tongue body region based on the color space data includes: obtaining the hue component and saturation component of the tongue body region in the color space data, and performing a difference operation with a preset healthy tongue color benchmark value to generate the tongue color feature value; obtaining the luminance component of the tongue body region in the color space data, and constructing a gray-level co-occurrence matrix of the luminance component to extract texture parameters from the gray-level co-occurrence matrix as the tongue coating texture feature value.

[0010] In one possible implementation of the first aspect, the step of generating a comprehensive efficacy index characterizing the overall efficacy of the periodic symptom-based quantitative data, the tongue color feature value, and the tongue coating texture feature value by pre-setting a combination includes: performing combination calculations on the periodic symptom-based quantitative data, the tongue color feature value, and the tongue coating texture feature value with pre-set first weights, second weights, and third weights respectively to obtain the comprehensive efficacy index.

[0011] In one possible implementation of the first aspect, the expression for the cumulative diffusion model of therapeutic efficacy is:

[0012]

[0013] Where N(t) is the cumulative comprehensive therapeutic improvement up to time t, m is the maximum potential therapeutic improvement, p is the therapeutic innovation coefficient, q is the therapeutic imitation coefficient, and t is the treatment cycle.

[0014] In one possible implementation of the first aspect, before fitting the time series of the comprehensive efficacy index to the efficacy cumulative diffusion model, the method further includes: acquiring the patient's constitution rhythm characteristic data and the seasonal data of the current treatment cycle; based on the constitution rhythm characteristic data and the seasonal data, indexing in a preset constitution seasonal response matrix to obtain a constitution seasonal correction factor; combining the initial efficacy innovation coefficient and efficacy imitation coefficient with the constitution seasonal correction factor to generate a corrected efficacy innovation coefficient and a corrected efficacy imitation coefficient; and using the corrected efficacy innovation coefficient and the corrected efficacy imitation coefficient as coefficients for fitting the efficacy cumulative diffusion model.

[0015] In one possible implementation of the first aspect, the body rhythm characteristic data is obtained by mapping the patient's birth date information to the sexagenary cycle system.

[0016] In one possible implementation of the first aspect, the step of comparing the predicted efficacy trend with a preset ideal recovery curve to generate a treatment plan adjustment signal includes: obtaining the difference between the predicted efficacy trend and the ideal recovery curve in a corresponding period; comparing the difference with a preset efficacy deviation threshold; and generating the treatment plan adjustment signal indicating adjustment if the difference exceeds the efficacy deviation threshold.

[0017] Secondly, this application provides a personalized traditional Chinese medicine enema efficacy dynamic optimization system, comprising: a data acquisition module, used to acquire periodic symptomatic quantitative data and periodic tongue image data characterizing the patient's current health status, as well as traditional Chinese medicine enema treatment plan data corresponding to the period; a tongue image feature extraction module, used to extract the tongue body region from the tongue image for each periodic tongue image data, convert the image data of the tongue body region to a preset color space to generate color space data, and based on the color space data, obtain the tongue color feature value and the tongue coating texture feature value of the tongue body region respectively; and an efficacy index generation module, used to, for each period, convert the periodic... The system uses a pre-defined weighted combination of symptomatic quantitative data, tongue color feature values, and tongue coating texture feature values ​​to generate a comprehensive efficacy index characterizing the overall efficacy of the treatment cycle. A efficacy trend prediction module is used to construct a time series of the comprehensive efficacy index based on the comprehensive efficacy index from multiple cycles. This time series is then fitted to a pre-defined efficacy cumulative diffusion model to determine the model's efficacy innovation coefficient and efficacy imitation coefficient. These coefficients are then substituted into the model to generate a predicted efficacy trend for one or more future cycles. An adjustment signal generation module compares the predicted efficacy trend with a pre-defined ideal recovery curve to generate a treatment plan adjustment signal indicating adjustments to the traditional Chinese medicine enema treatment plan.

[0018] In one possible implementation of the second aspect, the efficacy trend prediction module is further configured to: acquire the patient's constitution rhythm characteristic data and the seasonal data of the current treatment cycle; based on the constitution rhythm characteristic data and the seasonal data, index in a preset constitution seasonal response matrix to obtain a constitution seasonal correction factor; combine the initial efficacy innovation coefficient and efficacy imitation coefficient with the constitution seasonal correction factor to generate a corrected efficacy innovation coefficient and a corrected efficacy imitation coefficient; and use the corrected efficacy innovation coefficient and the corrected efficacy imitation coefficient as coefficients for fitting the efficacy cumulative diffusion model.

[0019] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in any one of the first aspects. Attached Figure Description

[0020] Figure 1 A flowchart illustrating the dynamic optimization method provided for some embodiments of this application;

[0021] Figure 2 A flowchart illustrating the dynamic optimization method provided for some embodiments of this application;

[0022] Figure 3A schematic diagram of the structure of a dynamic optimization system provided for some embodiments of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0024] In the following description, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0025] Furthermore, in this application, directional terms such as "upper," "lower," "left," and "right" may be defined relative to the orientation of the components shown in the accompanying drawings. It should be understood that these directional terms can be relative concepts, used for relative description and clarification, and may change accordingly depending on the orientation of the components in the accompanying drawings.

[0026] In this application, unless otherwise expressly specified and limited, the term "connection" should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral part; it can be a direct connection or an indirect connection through an intermediate medium. Furthermore, the term "electrical connection" can refer to the manner in which an electrical connection is used to achieve signal transmission.

[0027] As used herein, “about,” “approximately,” or “approximately” includes the stated value and a reference value within an acceptable range of deviation from the given value, characterized in that the acceptable range of deviation is determined by a person skilled in the art taking into account the measurement under discussion and the error associated with the measurement of the given quantity (i.e., the limitations of the measurement method).

[0028] To make the objectives, technical solutions, and advantages of the present invention clearer, a detailed description of the embodiments of the present invention will be provided below. However, those skilled in the art will understand that, in order to keep the description concise and clear, detailed descriptions of some well-known components and processes have been omitted in the embodiments of the present invention.

[0029] This invention provides a personalized Chinese medicine enema efficacy dynamic optimization system, method, and storage medium.

[0030] In a specific application scenario, suppose a patient receives traditional Chinese medicine retention enema treatment for a certain disease (such as ulcerative colitis), with the treatment cycle set to once daily. The system will dynamically evaluate and optimize the treatment plan based on this treatment cycle (day).

[0031] like Figure 1 As shown, the method includes:

[0032] S101. Obtain periodic symptomatic quantitative data and periodic tongue image data that characterize the patient's current health status, as well as data on traditional Chinese medicine enema treatment plans corresponding to the period.

[0033] Periodic symptomatic quantitative data is a quantitative description of the patient's subjective symptoms, obtained at the end of each treatment cycle through a pre-set questionnaire or scale, such as daily recorded symptom scores. Periodic tongue image data is an objective record of the patient's tongue appearance, acquired through standardized image acquisition equipment under fixed conditions (such as fixed time, lighting, and angle each day). Data on the traditional Chinese medicine enema treatment plan records the specific treatment content used in the current cycle, including the composition of the prescription, the dosage of each herb, the preparation method, and the method of use.

[0034] For example, daily symptom scores are collected via a patient-side app: abdominal pain (0-10 points), frequency of diarrhea (times / day), and degree of rectal bleeding (quantified as 0-3). These scores constitute periodic symptomatic quantitative data. Simultaneously, a color image of the patient's tongue is captured, e.g., in JPEG format, 1024x768 pixels. The treatment plan for the day is recorded as: "Prescription: Scutellaria baicalensis 10g, Coptis chinensis 6g, Rheum palmatum 5g; Decoction: Add 400ml of water, decoct to 200ml; Usage: Retention enema, 200ml each time, once daily."

[0035] S201. For each periodic tongue image data, extract the tongue region from the tongue image and convert the image data of the tongue region to a preset color space to generate color space data.

[0036] Image processing techniques, such as image segmentation algorithms based on color thresholding, edge detection, or machine learning models, are used to automatically or semi-automatically identify and extract the pixel set of the tongue region. The extracted image data of the tongue region is then converted from its original color space (such as RGB) to a preset color space, such as HSV (Hue, Saturation, Value), which separates color information (hue H, saturation S) from brightness information (value V), which is beneficial for subsequent feature extraction.

[0037] For example, a tongue region segmentation algorithm is performed on the acquired JPEG image of the tongue to obtain a binary mask of the same size as the image. In the mask, the pixel value of the tongue region is 1, and the background pixel value is 0. The RGB pixel values ​​of the corresponding tongue region in the original image are extracted using this mask. For each tongue pixel's RGB value (R, G, B), it is converted using a standard RGB to HSV conversion formula to obtain the corresponding HSV value (H, S, V). The set of HSV values ​​for all tongue region pixels constitutes the color space data.

[0038] S301. Based on the color space data, obtain the tongue color feature value and the tongue coating texture feature value for the tongue body region, respectively. The tongue color feature value is used to reflect the deviation of the tongue color from the health benchmark. The tongue coating texture feature value is used to reflect the visual characteristics of the tongue coating, such as roughness and thickness.

[0039] For example, when obtaining tongue color feature values, the average hue of all pixels in the tongue region color space data is calculated. and average saturation The average hue of a healthy tongue is obtained from the statistical analysis of tongue images of a pre-set group of healthy individuals. and the average saturation of the healthy tongue color benchmark Tongue color characteristic value can be calculated as the Euclidean distance between the current average tongue color and the healthy baseline color on the HS plane:

[0040]

[0041] This value quantifies the degree of deviation between the current tongue color and the healthy tongue color. When obtaining tongue texture feature values, a grayscale image is constructed using the luminance component V from the tongue region's color space data. The grayscale co-occurrence matrix (GLCM) of this grayscale image is calculated. Texture parameters, such as contrast, energy, and homogeneity, are extracted from the GLCM. The texture parameters extracted from the GLCM are selected as tongue texture feature values. For example, contrast is chosen as the tongue texture feature value, reflecting the intensity of local grayscale changes and the clarity of the texture.

[0042]

[0043] Where p(i,j) is the probability value of being located at (i,j) in the normalized GLCM, and N g It refers to grayscale levels. A higher contrast value usually indicates a coarser texture, which may correspond to a thick, greasy, or peeling tongue coating.

[0044] S401. For each cycle, the periodic symptom-based quantitative data, the tongue color feature value, and the tongue coating texture feature value are pre-combined to generate a comprehensive efficacy index that characterizes the overall efficacy of the cycle.

[0045] Quantitative indicators from different sources are integrated into a single Comprehensive Efficacy Index (CEI) to comprehensively assess the treatment effect of the current cycle. Before combination, each indicator is usually normalized to ensure it falls within the same numerical range (e.g., 0 to 1) to eliminate dimensional differences. Preset first weight w1, second weight w2, and third weight w3 are set by clinical experts based on experience or learned from historical data using machine learning methods. These reflect the relative importance of each indicator in assessing overall efficacy and satisfy w1 + w2 + w3 = 1.

[0046] For example, the normalized symptom-based quantitative data D norm Tongue color characteristic value C norm Tongue coating texture feature value T norm The weighted summation yields the comprehensive efficacy index (CEI):

[0047] CEI=w1×D norm +w2×C norm +w3×T norm

[0048] For example, if the normalized symptomatic data for a certain period is 0.75, the normalized tongue color feature value is 0.50, and the normalized tongue coating texture feature value is 0.60, and the weights are set to w1 = 0.6, w2 = 0.25, and w3 = 0.15, then the CEI for that period is 0.6 × 0.75 + 0.25 × 0.50 + 0.15 × 0.60 = 0.45 + 0.125 + 0.09 = 0.665. A higher CEI value can be set to indicate a more severe condition or a less ideal treatment outcome.

[0049] S501. Based on the comprehensive efficacy indicators of multiple periods, construct a time series of comprehensive efficacy indicators.

[0050] The comprehensive efficacy indicators obtained from consecutive treatment cycles are arranged in chronological order to form a sequence, which serves as the input for subsequent trend prediction models.

[0051] For example, if the system has recorded the CEI values ​​of the patient for the first 10 treatment cycles, namely {CEI1, CEI2, ..., CEI...} 10 If we construct a time series, then the constructed time series is the ordered set.

[0052] S601. Fit the time series of the comprehensive efficacy index to a preset efficacy cumulative diffusion model to determine the efficacy innovation coefficient and efficacy imitation coefficient of the model.

[0053] In this application, the therapeutic effect is considered a "diffusion" process, where the direct action of the drug is similar to an "innovation" effect, and the body's systemic and sustained response to the drug and its own recovery ability are similar to an "imitation" effect. The cumulative function is in the form of:

[0054]

[0055] Where t represents the number of treatment cycles, and N(t) represents the cumulative improvement in overall efficacy up to cycle t. If the initial (pre-treatment) overall efficacy index is set as CEI... initial Furthermore, the lower the CEI value, the better the therapeutic effect; therefore, the cumulative improvement N(t) = CEI. initial -CEI t m is the maximum potential therapeutic improvement, which can theoretically be set as CEI. initial (Corresponding to a CEI of 0). p is the efficacy innovation coefficient, quantifying the drug's rapid onset of action. q is the efficacy imitation coefficient, quantifying the body's systemic and sustained response to treatment.

[0056] The time series {CEI1,...,CEI n} is converted into a cumulative improvement sequence {N(1),...,N(n)}, where N(t)=CEI initial -CEI t Using optimization algorithms such as nonlinear least squares, these (t, N(t)) data points are fitted to the above model to solve for the p-value and q-value that best describe the current dynamics of the therapeutic effect.

[0057] For example, obtain the CEI values ​​{CEI1,...,CEI} for the first 15 periods. 15}, calculate the corresponding cumulative improvement amount {N(1),...,N} 15}, where CEI initial These are the CEI values ​​measured before treatment begins. Using these data points, the above model is fitted using a nonlinear regression algorithm (such as the Levenberg-Marquardt algorithm) to determine the optimal p-value and q-value. For example, the fitting result might be p = 0.08, q = 0.35.

[0058] S701. Substitute the efficacy innovation coefficient and the efficacy imitation coefficient into the efficacy cumulative diffusion model to generate a predicted efficacy trend for one or more future periods.

[0059] Using a defined efficacy innovation coefficient p and efficacy imitation coefficient q, and the maximum potential efficacy improvement m (e.g., set as CEI) initialSubstituting these values ​​into the diffusion model formula, we calculate the predicted cumulative improvement in efficacy at future time points t' (e.g., n+1, n+2, ..., n+k after the current cycle number n). Then, we convert the predicted cumulative improvement back into the predicted comprehensive efficacy index, Predicted_CEI. t′ =CEI initial -N(t ′ This allows us to obtain predictions of therapeutic trends for future cycles.

[0060] For example, using the p = 0.08 and q = 0.35 obtained above, set m = CEI initial =0.90. Predicted cumulative improvement for the next 16 and 17 days:

[0061]

[0062] The corresponding predictive comprehensive efficacy index is: Predicted_CEI 16 =0.90-0.895=0.005, Predicted_CEI 17 =0.90-0.897=0.003.

[0063] S801. Compare the predicted therapeutic trend with the preset ideal recovery curve to generate a treatment plan adjustment signal for indicating adjustments to the traditional Chinese medicine enema treatment plan.

[0064] The pre-defined ideal recovery curve is a path of expected overall efficacy index (CEI) decline over treatment time, set by clinical experts based on their understanding of the specific disease and patient condition. The predicted future CEI value is compared with the expected CEI value at the corresponding time point on the ideal recovery curve.

[0065] For example, an ideal recovery curve is preset, such as the expected CEI to decrease from the initial value of 0.90 to 0.10 within 30 days. Its functional expression is Ideal_CEI(t) = 0.90 - (0.90 - 0.10) / 30 × t = 0.90 - 0.0267 × t. The predicted CEI on the 16th and 17th days in the future is compared with the ideal value:

[0066] Ideal CEI 16 =0.90 - 0.0267 × 16 = 0.90 - 0.4272 = 0.4728

[0067] Ideal CEI 17 =0.90 - 0.0267 × 17 = 0.90 - 0.4539 = 0.4461

[0068] Calculate the difference ΔCEI between the predicted value and the ideal value. t=Predicted_CEI t -Ideal_CEI t Set a threshold for the therapeutic effect to deviate from the threshold, for example, set it to 0.05.

[0069] For day 16, ΔCEI 16 =0.005-0.4728=-0.4678. Since |-0.4678|>0.05 and the difference is negative (the predicted efficacy far exceeds expectations), the system generates a treatment plan adjustment signal of "suggesting a slowdown in treatment".

[0070] For day 17, ΔCEI 17 =0.003-0.4461=-0.4431. Similarly, since |-0.4431|>0.05 and the difference is negative, the system generates a treatment plan adjustment signal of "suggesting a slowdown in treatment".

[0071] If the predicted value is slightly higher than the ideal value but the absolute value of the difference is less than or equal to the threshold, for example, Predicted_CEI 16 =0.50 (Δ = 0.50 - 0.4728 = 0.0272 ≤ 0.05), then the "maintain current plan" signal is generated.

[0072] If the predicted value is significantly higher than the ideal value, for example, Predicted_CEI 16 =0.60 (Δ = 0.60 - 0.4728 = 0.1272 > 0.05), then a "recommendation to strengthen treatment" signal is generated.

[0073] To enhance the personalization and accuracy of efficacy prediction models, a constitution-seasonal correction mechanism can be introduced before fitting the comprehensive efficacy index time series to the efficacy cumulative diffusion model.

[0074] like Figure 2 As shown, before executing S601, the following may also be included:

[0075] S901. Obtain the patient's physical rhythm characteristics data and the seasonal data of the current treatment cycle.

[0076] Factors related to the patient's individual constitution and current natural environment are obtained. Constitutional rhythm characteristics data can be obtained by converting the patient's birth date (Gregorian calendar) to the lunar calendar and mapping it to a sexagenary cycle system (such as the sexagenary cycle). The sexagenary cycle is closely related to the Five Elements and Six Qi theory in Traditional Chinese Medicine, reflecting an individual's innate constitutional bias. Seasonal data is obtained by acquiring the current treatment date and querying the corresponding one of the twenty-four solar terms, reflecting the current natural climate and seasonal changes.

[0077] For example, the patient's date of birth is obtained as May 10, 1990 in the Gregorian calendar. Converted to the lunar calendar, this is the 16th day of the fourth month of the Gengwu year. Using the sexagenary cycle conversion, the birth year is determined to be the Gengwu year. The constitution rhythm characteristic data is "Gengwu". The current treatment date is obtained as June 15, 2023 in the Gregorian calendar. Querying the twenty-four solar terms table, this date falls within the "Mangzhong" solar term. The solar term seasonal data is "Mangzhong".

[0078] S1001. Based on the physical rhythm characteristic data and the solar term data, index the preset physical seasonal response matrix to obtain the physical seasonal correction factor.

[0079] The pre-defined constitution-seasonal response matrix is ​​a knowledge lookup table or database that records the modification factors of different constitution types (e.g., represented by the sexagenary cycle) on the therapeutic process (especially the accumulation and persistence of therapeutic effects, i.e., the imitation effect) under different solar terms. This matrix can be constructed by those skilled in the art based on classical Chinese medicine theories (such as the Five Elements and Six Qi chapter of the *Huangdi Neijing*) and modern clinical research.

[0080] For example, this matrix can be a two-dimensional table, with row indices representing the sexagenary cycle (60 types) and column indices representing the twenty-four solar terms (24 types). Each element of the matrix, `Correction_Factor(Constitution, Season)`, is a numerical value. For instance, the constitution "Gengwu" and the season "Mangzhong" obtained through S901 are indexed in this matrix to the corresponding correction factor, such as `Correction_Factor('Gengwu', 'Mangzhong') = 1.15`. This correction factor indicates that people with a Gengwu constitution may experience a 15% greater mimicry effect on treatment during the Mangzhong solar term than the average.

[0081] S1101. The initial therapeutic innovation coefficient and the therapeutic imitation coefficient are combined with the constitution-seasonal correction factor to generate the corrected therapeutic innovation coefficient and the corrected therapeutic imitation coefficient.

[0082] The efficacy innovation coefficient p obtained from the preliminary fitting based on historical data initial And the efficacy imitation coefficient q initial This is combined with a constitution-seasonal correction factor. Generally, constitution and seasonal factors have a more direct impact on the persistence of therapeutic effects (imitation effect q). In some embodiments, the combination can be multiplicative.

[0083] For example, the correction factor is applied to the imitation coefficient:

[0084] p corrected =p initial

[0085] q corrected =qinitial ×Correction_Factor

[0086] If p is obtained through preliminary fitting based on historical data initial =0.08,q initial =0.35, and the correction factor is 1.15, then the corrected efficacy imitation coefficient is q. corrected =0.35×1.15=0.4025.

[0087] S1201. The modified efficacy innovation coefficient and the efficacy imitation coefficient are used as coefficients to fit the efficacy cumulative diffusion model.

[0088] The revised efficacy innovation coefficient p corrected And the efficacy imitation coefficient q corrected This is used in the prediction process. That is, in S701, the corrected coefficients are substituted into the prediction model to predict future efficacy trends.

[0089]

[0090] In this way, by incorporating the patient's physical constitution and current natural environmental factors into the efficacy prediction model, the prediction results are more in line with the holistic view of "harmony between man and nature" in traditional Chinese medicine, thus improving the personalization and accuracy of the prediction.

[0091] This invention also provides a personalized traditional Chinese medicine enema efficacy dynamic optimization system 100. This system 100 is a computing system that can be deployed on a server, cloud platform, or dedicated device, and interacts with data acquisition devices (such as patient mobile apps) and decision support terminals (such as physician workstations) via network interfaces. Figure 3 As shown, system 100 may include:

[0092] The data acquisition module 110 is responsible for receiving, storing, and managing periodic symptom-based quantitative data, periodic tongue image data, and traditional Chinese medicine enema treatment plan data from various channels. This module should ensure the integrity, accuracy, and timestamp of the data.

[0093] The tongue image feature extraction module 120 receives tongue image data and contains an image processing algorithm unit for performing tongue region segmentation, color space conversion, and further calculating the tongue color feature value and tongue coating texture feature value of the tongue region.

[0094] The efficacy index generation module 130 receives periodic symptom-based quantitative data, tongue color characteristic values, and tongue coating texture characteristic values. This module contains a data normalization unit and a weighted combination calculation unit, which calculates and generates periodic comprehensive efficacy indicators based on preset weights.

[0095] The efficacy trend prediction module 140 is the core calculation module of system 100. It obtains the comprehensive efficacy index sequence from the historical period from the efficacy index generation module 130 and constructs a time series. This module embeds a fitting unit for the cumulative efficacy diffusion model, used to fit historical data and determine the model's efficacy innovation coefficient p and efficacy imitation coefficient q. After fitting, this module uses the determined coefficients to predict the cumulative efficacy improvement for one or more future periods and converts it into a predicted comprehensive efficacy index sequence, forming a predicted efficacy trend.

[0096] The module may also optionally include a constitution-seasonal correction unit, which interacts with the data acquisition module 110 to acquire constitution rhythm characteristic data and solar term seasonal data. It contains a constitution-seasonal response matrix knowledge base, which is used to query and obtain constitution-seasonal correction factors, and apply the correction factors to the efficacy innovation coefficient and / or efficacy imitation coefficient, and use the corrected coefficients for model fitting or prediction.

[0097] The adjustment signal generation module 150 receives the predicted efficacy trend output by the efficacy trend prediction module 140 and retrieves the ideal recovery curve data from a preset storage unit. This module performs a comparison calculation between the predicted trend and the ideal curve (e.g., calculating the difference between the predicted value and the ideal value) and compares the difference with a preset efficacy deviation threshold. Based on the comparison result, a structured treatment plan adjustment signal is generated. This signal can be sent to an external system via an interface for display or notification.

[0098] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by one or more processors, it can implement the method for dynamically optimizing the efficacy of personalized traditional Chinese medicine enema as described in any of the preceding claims. The computer program may include instruction code for implementing the above-described method. The computer-readable storage medium may be any type of non-transitory computer-readable storage medium, including but not limited to: ROM, RAM, magnetic recording media (e.g., hard disk, floppy disk, magnetic tape), optical recording media (e.g., CD-ROM, DVD, Blu-ray disc), flash memory, etc. When the computer program is run on a processor, the processor reads and executes the instructions in the program, thereby enabling the processor to implement the functions and methods described in the method of the present invention.

[0099] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0100] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0101] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or 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 apparatus, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0102] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units; that is, it can be located in one place or distributed in multiple different locations. Depending on actual needs, some or all of the units can be selected to achieve the purpose of this embodiment.

[0103] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit described above can be implemented in hardware.

[0104] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope 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 method for dynamically optimizing the efficacy of personalized traditional Chinese medicine enema, characterized in that, include: Acquire periodic symptomatic quantitative data and periodic tongue image data that characterize the patient's current health status, as well as data on traditional Chinese medicine enema treatment plans corresponding to the period; For each periodic tongue image data, the tongue region in the tongue image is extracted, and the image data of the tongue region is converted to a preset color space to generate color space data; Based on the color space data, the tongue color feature value and tongue coating texture feature value of the tongue body region are obtained respectively; For each cycle, the periodic symptom-based quantitative data, the tongue color feature value, and the tongue coating texture feature value are pre-combined to generate a comprehensive efficacy index that characterizes the overall efficacy of that cycle. Based on the comprehensive efficacy indicators over multiple periods, a time series of comprehensive efficacy indicators is constructed; The time series of the comprehensive efficacy indicators are fitted to a preset efficacy cumulative diffusion model to determine the efficacy innovation coefficient and efficacy imitation coefficient of the model; Substituting the efficacy innovation coefficient and the efficacy imitation coefficient into the efficacy cumulative diffusion model, a predicted efficacy trend for one or more future periods is generated; The predicted therapeutic trend is compared with the preset ideal recovery curve to generate a treatment plan adjustment signal for indicating adjustments to the traditional Chinese medicine enema treatment plan.

2. The method according to claim 1, characterized in that, The step of obtaining the tongue color feature value and the tongue coating texture feature value of the tongue body region based on the color space data includes: In the color space data, the hue component and saturation component of the tongue region are obtained, and the difference is calculated with the preset healthy tongue color benchmark value to generate the tongue color feature value; In the color space data, the luminance component of the tongue region is obtained, and a gray-level co-occurrence matrix of the luminance component is constructed to extract texture parameters from the gray-level co-occurrence matrix as the texture feature values ​​of the tongue coating.

3. The method according to claim 1, characterized in that, The step of generating a comprehensive efficacy index characterizing the overall efficacy of the periodic symptomatic quantitative data, the tongue color feature value, and the tongue coating texture feature value by pre-setting a combination includes: The periodic symptom-based quantitative data, the tongue color feature value, and the tongue coating texture feature value are combined with preset first weights, second weights, and third weights to obtain the comprehensive therapeutic effect index.

4. The method according to claim 1, characterized in that, The expression for the cumulative diffusion model of therapeutic efficacy is: Where N(t) is the cumulative comprehensive therapeutic improvement up to time t, m is the maximum potential therapeutic improvement, p is the therapeutic innovation coefficient, q is the therapeutic imitation coefficient, and t is the treatment cycle.

5. The method according to claim 1 or 4, characterized in that, Before fitting the time series of the comprehensive efficacy index to the cumulative diffusion model of efficacy, the method further includes: Obtain patient's physical rhythm characteristics data and the seasonal data of the current treatment cycle; Based on the physical rhythm characteristic data and the solar term data, an index is performed in the preset physical seasonal response matrix to obtain the physical seasonal correction factor. The initial therapeutic innovation coefficient and the therapeutic imitation coefficient are combined with the constitution-seasonal correction factor to generate the corrected therapeutic innovation coefficient and the corrected therapeutic imitation coefficient. The modified efficacy innovation coefficient and the modified efficacy imitation coefficient are used as coefficients to fit the efficacy cumulative diffusion model.

6. The method according to claim 5, characterized in that, The physical rhythm characteristic data is obtained by mapping the patient's birth date information to the sexagenary cycle system.

7. The method according to claim 1, characterized in that, The step of comparing the predicted therapeutic trend with a preset ideal recovery curve to generate a treatment plan adjustment signal includes: Obtain the difference between the predicted therapeutic trend and the ideal recovery curve at the corresponding cycle; The difference is compared with a preset threshold for deviation of therapeutic effect; If the difference exceeds the efficacy deviation threshold, an adjustment signal indicating adjustment of the treatment plan is generated.

8. A personalized traditional Chinese medicine enema efficacy dynamic optimization system, characterized in that, include: The data acquisition module is used to acquire periodic symptomatic quantitative data and periodic tongue image data that characterize the patient's current health status, as well as data on traditional Chinese medicine enema treatment plans corresponding to the period. The tongue image feature extraction module is used to extract the tongue body region from each periodic tongue image image data, convert the image data of the tongue body region to a preset color space to generate color space data, and obtain the tongue color feature value and the tongue coating texture feature value of the tongue body region based on the color space data. The efficacy index generation module is used to generate a comprehensive efficacy index that characterizes the overall efficacy of the periodic symptom quantitative data, the tongue color feature value, and the tongue coating texture feature value for each period by performing a preset weighted combination. The efficacy trend prediction module is used to construct a time series of comprehensive efficacy indicators based on the comprehensive efficacy indicators of multiple periods, fit the time series to a preset efficacy cumulative diffusion model to determine the efficacy innovation coefficient and efficacy imitation coefficient of the model, and substitute the coefficients into the model to generate a predicted efficacy trend for one or more future periods. The adjustment signal generation module is used to compare the predicted therapeutic trend with the preset ideal recovery curve and generate a treatment plan adjustment signal to indicate the adjustment of the traditional Chinese medicine enema treatment plan.

9. The system according to claim 8, characterized in that, The efficacy trend prediction module is also used for: Obtain patient's physical rhythm characteristics data and the seasonal data of the current treatment cycle; Based on the physical rhythm characteristic data and the solar term data, an index is performed in the preset physical seasonal response matrix to obtain the physical seasonal correction factor. The initial therapeutic innovation coefficient and the therapeutic imitation coefficient are combined with the constitution-seasonal correction factor to generate the corrected therapeutic innovation coefficient and the corrected therapeutic imitation coefficient. The modified efficacy innovation coefficient and the modified efficacy imitation coefficient are used as coefficients to fit the efficacy cumulative diffusion model.

10. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in any one of claims 1-7.

Images