CN-122004770-A - Method and system for detecting fruits with external curative effects of traditional Chinese medicine based on multiple modes

CN122004770ACN 122004770 ACN122004770 ACN 122004770ACN-122004770-A

Abstract

The invention relates to a multi-mode-based detection method and system for fruits with external curative effects of traditional Chinese medicines, and belongs to the technical field of medical detection. The method comprises the steps of obtaining infrared thermal imaging, laser speckle blood perfusion, skin conductance and body surface temperature signals of a traditional Chinese medicine external application area of a subject, inputting a multi-mode fusion model to generate fusion feature vectors after time synchronization, noise suppression and spatial registration, calling preset physiological response reference data, respectively comparing the fusion feature vectors with the two to generate first comparison result data and second comparison result data, generating physiological response intensity data according to the first comparison result data and the second comparison result data, inputting the intensity data of a continuous period into a trend prediction model, outputting physiological response trend data in a future time window, and realizing objective, real-time, quantitative and predictive monitoring of local physiological state change under the action of external application of the traditional Chinese medicine through multi-dimensional signal fusion and double-reference dynamic comparison.

Inventors

YUAN XINZHOU
YUAN JIAFENG
CHEN YONGDE

Assignees

深圳市鑫舟江领生物信息科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260224

Claims (10)

1. The detection method for fruits with external curative effects of traditional Chinese medicine based on multiple modes is characterized by comprising the following steps of: acquiring a multidimensional physiological signal of a traditional Chinese medicine external application area of a subject, wherein the multidimensional physiological signal comprises an infrared thermal imaging signal, a laser speckle blood perfusion signal, a skin conductance signal and a body surface temperature signal; Performing time synchronization, noise suppression and spatial registration processing on the multidimensional physiological signals to obtain preprocessed multidimensional physiological signals; Inputting the preprocessed multidimensional physiological signals into a multi-modal fusion model, and outputting fusion feature vectors representing local physiological states by the multi-modal fusion model; According to preset physiological response reference data, the physiological response reference data comprise first reference data and second reference data, wherein the first reference data are multi-mode physiological baseline feature vectors of a healthy subject corresponding to a body surface area in a state of not applying traditional Chinese medicine external application, and the second reference data are fusion feature vectors of the same subject at the initial moment of traditional Chinese medicine external application; comparing the fusion feature vector with the first reference data to generate first comparison result data, wherein the first comparison result data represents the deviation degree of the current physiological state relative to a group baseline; Comparing the fusion feature vector with the second reference data to generate second comparison result data, wherein the second comparison result data represents the change amplitude of the current physiological state relative to the initial state of the individual; generating physiological response intensity data based on the first comparison result data and the second comparison result data; and inputting the physiological response intensity data generated in the continuous time period into a trend prediction model, and outputting physiological response trend data in a future preset time window by the trend prediction model.
2. The method for detecting fruits on external curative effect of multi-modal based traditional Chinese medicine according to claim 1, wherein before inputting the preprocessed multi-dimensional physiological signals into the multi-modal fusion model, further comprising: Identifying a pharmacodynamic diffusion region based on the infrared thermal imaging signal and generating pharmacodynamic diffusion region data; Identifying a conductance abnormal region corresponding to a preset meridian path or acupoint position based on the skin conductance signal, and generating meridian response characteristic data; and taking the drug effect diffusion area data, the microcirculation response mode data and the meridian response characteristic data as auxiliary inputs, and inputting the auxiliary inputs and the preprocessed multidimensional physiological signals into the multi-mode fusion model.
3. The method for detecting fruits with external curative effects based on multimode traditional Chinese medicines according to claim 1, wherein the multimode fusion model is configured based on preset medicine permeation parameters, and the medicine permeation parameters comprise component information, effective component concentration and application duration of the external application preparation of the traditional Chinese medicines; The multi-modal fusion model generates a fusion feature vector for representing the coupling characteristic between the drug action and the physiological response according to the association relationship between the drug permeation parameter and the preprocessed multi-dimensional physiological signal.
4. The method for detecting fruits with external curative effects based on multi-mode traditional Chinese medicines according to claim 1, wherein the step of comparing the fusion feature vector with the first reference data to generate first comparison data comprises the following steps: calculating the mahalanobis distance between the fusion feature vector and the first reference data as the first comparison data; The step of comparing the fusion feature vector with second reference data to generate second comparison result data comprises the following steps: And calculating cosine similarity between the fusion feature vector and the second reference data to be used as the second comparison result data.
5. The method for detecting fruits with external curative effects based on multi-modal traditional Chinese medicines according to claim 1, wherein generating physiological response intensity data based on the first comparison result data and the second comparison result data comprises: calculating the physiological response intensity data by adopting a weighted fusion formula : ; Wherein, the The mahalanobis distance characterized for the first comparison data, Is a preset maximum allowable mahalanobis distance threshold, Cosine similarity characterized for the second comparison result data, The value range is that the weight coefficient is For balancing the relative importance of population baseline wander versus individual state changes.
6. The method for detecting fruits with external curative effects based on multi-mode traditional Chinese medicines according to claim 1, wherein the generation process of the physiological response intensity data comprises combining with traditional Chinese medicine dialectical rule data, and the traditional Chinese medicine dialectical rule data comprises: Based on the cold and heat attribute discrimination rule corresponding to the spatial distribution characteristic of the thermal imaging signal; based on a qi-blood unobstructed degree discrimination rule corresponding to time sequence fluctuation characteristics of blood flow perfusion signals; And the physiological response intensity data is subjected to weighted adjustment according to the matching result of the fusion feature vector and the traditional Chinese medicine dialectical rule data.
7. The method for detecting fruits with external curative effects of multi-modal based on traditional Chinese medicines according to claim 4, further comprising the step of comparing the fusion feature vector with the first reference data to generate first comparison result data, and further comprising the steps of: The fusion feature vector is calculated as follows: ; the formula calculates the standardized Markov distance of the current fusion feature vector relative to the healthy population baseline multi-element distribution, and the larger the Z value is, the higher the comprehensive degree of the current local physiological state deviating from the population healthy baseline is.
8. The method for detecting fruits with external curative effects of multi-modal based Chinese medicine according to claim 7, further comprising the step of, after generating the physiological response intensity data, evaluating: comparing the physiological response intensity data with a preset intensity-efficacy association threshold; if the physiological response intensity data continuously exceeds a high response threshold value, generating a positive curative effect prompt signal; if the physiological response intensity data is continuously lower than the low response threshold value, generating a parameter adjustment suggestion signal, wherein the suggestion comprises adjusting the concentration of the external application formula or the application duration of the traditional Chinese medicine.
9. A multi-mode-based detection system for fruits with external curative effects of traditional Chinese medicine, which is used for the multi-mode-based detection method for fruits with external curative effects of traditional Chinese medicine according to any one of claims 1 to 8, and is characterized by comprising the following steps: The signal acquisition module is used for acquiring multidimensional physiological signals of the traditional Chinese medicine external application area of the subject, wherein the multidimensional physiological signals comprise an infrared thermal imaging signal, a laser speckle blood flow perfusion signal, a skin conductance signal and a body surface temperature signal; The preprocessing module is used for carrying out time synchronization, noise suppression and spatial registration processing on the multidimensional physiological signals to obtain preprocessed multidimensional physiological signals; the fusion processing module is used for inputting the preprocessed multidimensional physiological signals into a multi-modal fusion model, and outputting fusion feature vectors representing local physiological states by the multi-modal fusion model; The reference basis and comparison module is used for comparing the reference data according to preset physiological response, wherein the physiological response reference data comprises first reference data and second reference data, the first reference data is a multi-mode physiological baseline characteristic vector of a healthy subject corresponding to a body surface area in a state of not applying traditional Chinese medicine external application, and the second reference data is a fusion characteristic vector of the same subject at the initial moment of traditional Chinese medicine external application; the reference is also used for comparing the fusion feature vector with the first reference data according to the comparison module to generate first comparison data, and comparing the fusion feature vector with the second reference data to generate second comparison data; The response intensity generation module is used for generating physiological response intensity data based on the first comparison result data and the second comparison result data; The trend prediction module is used for inputting the physiological response intensity data generated in the continuous time period into a trend prediction model, and outputting physiological response trend data in a future preset time window by the trend prediction model; the output module is used for outputting the physiological response intensity data and the physiological response trend data and objectively recording the dynamic change process of the local physiological state of the human body under the action of external application of the traditional Chinese medicine.
10. A computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the multi-modality based method of detecting fruits of chinese medicine external curative effects of any of claims 1 to 8.

Description

Method and system for detecting fruits with external curative effects of traditional Chinese medicine based on multiple modes Technical Field The invention belongs to the technical field of medical detection, and particularly relates to a multi-mode-based detection method for fruits with external curative effects of traditional Chinese medicines. Background The external application of traditional Chinese medicine is a traditional application mode, and a series of complex physiological state changes can be caused after the traditional Chinese medicine acts on the skin area of a human body. At present, an objective and quantitative monitoring and analyzing means is not available for the physiological responses caused by external application of traditional Chinese medicines. The traditional observation mode mainly depends on subjective experience judgment of operators and self-perception description of a subject, and has obvious technical limitations that firstly, an evaluation result highly depends on personal experiences of the operators, and large differences can exist among judgment results of different operators, so that evaluation lacks consistency and comparability, secondly, the traditional method cannot continuously and timely monitor dynamic evolution of human physiological states in the traditional Chinese medicine external application process, generally can only carry out one-time and hysteresis summarization after application is finished, and thirdly, the traditional observation mode usually only focuses on single physiological indexes (such as body surface temperature or subjective comfort level) and is difficult to comprehensively and comprehensively describe physiological state changes from multiple dimensions such as blood circulation, metabolic activity, skin electrical characteristics and the like, and finally, objective prediction of physiological state change trend cannot be formed due to lack of deep mining and analysis of multidimensional physiological data. The above problems seriously hamper the deep understanding of the action mechanism of external application of the traditional Chinese medicine and the scientific and standardized research of the application process. Therefore, a technical scheme capable of objectively, real-time and multidimensional monitoring and analyzing the local physiological state change of a human body under the action of external application of the traditional Chinese medicine is needed. Disclosure of Invention The invention aims to provide a multi-mode traditional Chinese medicine external application physiological response real-time monitoring and analyzing method and system, which are used for solving the technical problems that in the prior art, the subjective evaluation of physiological state change caused by traditional Chinese medicine external application is strong, the real-time monitoring cannot be realized, the multi-dimensional data comprehensive analysis is lacked, and the physiological response trend cannot be objectively predicted. In order to achieve the above purpose, the first aspect of the present invention provides a method for detecting fruits with external curative effects based on multi-mode traditional Chinese medicine, comprising the following steps: acquiring a multidimensional physiological signal of a traditional Chinese medicine external application area of a subject, wherein the multidimensional physiological signal comprises an infrared thermal imaging signal, a laser speckle blood perfusion signal, a skin conductance signal and a body surface temperature signal; Performing time synchronization, noise suppression and spatial registration processing on the multidimensional physiological signals to obtain preprocessed multidimensional physiological signals; Inputting the preprocessed multidimensional physiological signals into a multi-modal fusion model, and outputting fusion feature vectors representing local physiological states by the multi-modal fusion model; According to preset physiological response reference data, the physiological response reference data comprise first reference data and second reference data, wherein the first reference data are multi-mode physiological baseline feature vectors of a healthy subject corresponding to a body surface area in a state of not applying traditional Chinese medicine external application, and the second reference data are fusion feature vectors of the same subject at the initial moment of traditional Chinese medicine external application; comparing the fusion feature vector with the first reference data to generate first comparison result data, wherein the first comparison result data represents the deviation degree of the current physiological state relative to a group baseline; Comparing the fusion feature vector with the second reference data to generate second comparison result data, wherein the second comparison result data represents the change amplitude of the current physiol