Search

CN-122017162-A - Medicine quality detection method and system for treating rhinitis

CN122017162ACN 122017162 ACN122017162 ACN 122017162ACN-122017162-A

Abstract

The invention relates to the technical field of medicine quality detection, and discloses a medicine quality detection method and system for treating rhinitis. The method comprises establishing a quality characterization network comprising a drug matrix, an active ingredient distribution and a physical trait manifestation layer. And acquiring spectrum, chromatogram and granularity distribution data of the drug to be detected, and mapping the data to a network to generate an initial quality representation. And quantifying and calculating the association strength among the feature points to form a dynamic association link, and identifying the conduction path of the quality attribute and the key node according to the dynamic association link. Triggering network topology adjustment according to the path and the node state, and reassigning feature point weights. And performing quality judgment based on the updated network, and outputting a structured quality report. The method realizes the system quantitative analysis of dynamic association between the quality attributes of the medicines, can adaptively optimize the judgment logic according to the characteristics of the samples, and improves the depth and accuracy of quality assessment.

Inventors

  • XU YI

Assignees

  • 宁波市第二医院

Dates

Publication Date
20260512
Application Date
20260119

Claims (10)

  1. 1. A method for detecting the quality of a medicament for treating rhinitis, which is characterized by comprising the following steps: The method comprises the steps of establishing a quality characteristic network, wherein the quality characteristic network consists of a plurality of mutually related characteristic layers and is used for representing the multielement quality attribute of rhinitis drugs, and the quality characteristic network comprises a drug matrix layer, an active ingredient distribution layer and a physical property expression layer, wherein each layer consists of a plurality of quality characteristic points; Obtaining original sample data of rhinitis medicine to be detected, mapping the original sample data to corresponding feature points of the quality feature network, and generating an initial quality characterization graph, wherein the original sample data comprises spectrum data, chromatographic data and granularity distribution data; In the quality characteristic network, carrying out quantization calculation on the association strength between characteristic points in the initial quality characteristic diagram to form a dynamic association link; triggering topological structure adjustment of the quality feature network according to the states of the conduction path and the key nodes to generate an updated quality feature network; based on the quality characteristic network after weight redistribution, executing the quality judgment process of the rhinitis medicine to be detected, and outputting a structured quality report.
  2. 2. The method for detecting the quality of the medicine for treating rhinitis according to claim 1, wherein the process of establishing the quality characteristic network comprises the steps of extracting reference sample data from a standard rhinitis medicine sample library, carrying out cluster analysis on quality attributes in the reference sample data, identifying three core attribute clusters of medicine matrixes, active components and physical properties, defining each core attribute cluster as a characteristic layer, configuring independent characteristic coding rules and characteristic point association rules for each characteristic layer, wherein the characteristic coding rules are used for discretizing continuous quality data into state values of characteristic points, and the characteristic point association rules are used for defining potential connection relations of characteristic points in the same characteristic layer and among different characteristic layers.
  3. 3. The method for detecting the quality of the rhinitis treatment medicine according to claim 2, wherein the method is characterized by obtaining original sample data of the rhinitis treatment medicine to be detected, mapping the original sample data to corresponding characteristic points of the quality characteristic network to generate an initial quality characterization graph, specifically, invoking characteristic coding rules respectively corresponding to a medicine matrix layer, an active ingredient distribution layer and a physical property expression layer, carrying out parallel processing on the spectrum data, the spectrum data and the particle size distribution data, mapping the processed spectrum data into characteristic point state values of auxiliary materials in the medicine matrix layer, relating to the composition and uniformity of auxiliary materials, mapping the processed spectrum data into characteristic point state values of active ingredient distribution layer, relating to the content of main ingredients and impurity mass spectrum, mapping the processed particle size distribution data into characteristic point state values of physical property expression layer, relating to the particle morphology and particle size range, and integrating all the characteristic point state values and the position information of the characteristic point state values in the respective characteristic layers to form the initial quality characterization graph.
  4. 4. A drug quality detection method for treating rhinitis according to claim 3 is characterized in that in the quality feature network, the quantitative calculation is carried out on the correlation intensity among feature points in the initial quality feature graph to form a dynamic correlation link, the method comprises traversing all feature point pairs in the initial quality feature graph according to the feature point correlation rule, calculating a cooperative variation index of a state value and a distance attenuation coefficient of each feature point pair, multiplying the cooperative variation index by the distance attenuation coefficient to obtain initial correlation intensity of the feature point pair, continuously acquiring a plurality of batches of original sample data in a preset time window, repeating the mapping process to generate a series of quality feature graphs in time sequence, calculating variance of the correlation intensity of the same feature point pair in the time window, defining reciprocal of the variance as a correlation stability factor, and multiplying the initial correlation intensity by the correlation stability factor to obtain final correlation intensity of the dynamic correlation link.
  5. 5. The method for detecting the quality of medicines for treating rhinitis according to claim 4 is characterized by identifying conducting paths and key nodes of quality attributes in a network according to the strength change of the dynamic association links, specifically, setting an association strength threshold, screening out links with final association strength exceeding the association strength threshold in the dynamic association links, marking the links as effective association links, searching shortest paths among all feature point pairs in a sub-network formed by all feature points and the effective association links, counting the number of shortest paths passing through each feature point, defining the number of the shortest paths as the path centrality of the feature points, identifying the feature points with the path centrality ranked in front of a preset proportion as the key nodes, and defining a network structure formed by the key nodes and the effective association links between the key nodes as the main conducting paths.
  6. 6. The method for detecting the quality of medicines for treating rhinitis according to claim 5, wherein the step of triggering the topological structure adjustment of the quality feature network according to the states of the conducting path and the key nodes comprises the steps of monitoring the fluctuation frequency and the fluctuation amplitude of the state value of the key nodes on the conducting path, judging that the local network where a specific key node is located is in an unstable state when the fluctuation frequency of the state value of the specific key node exceeds a frequency threshold value and the fluctuation amplitude exceeds an amplitude threshold value, triggering the re-evaluation of all feature point associated links directly connected with the specific key node, wherein the re-evaluation process comprises the steps of temporarily freezing the state value of the key node, recalculating the cooperative change index and the distance attenuation coefficient between the key node and all adjacent feature points based on the latest batch of original sample data, generating new association strength, updating the weight of the feature point associated links and recording a topology change event if the difference between the new association strength and the original final association strength exceeds an adjustment threshold value.
  7. 7. The method for detecting the quality of a medicament for treating rhinitis according to claim 6, wherein the method for generating an updated quality feature network is characterized by comprising the steps of carrying out reassignment on weights of feature points in the updated quality feature network, counting total number and distribution of topology change events caused by all key nodes in a preset period, calculating the number of times that each feature point is influenced by the topology change event according to the distribution of the topology change events, carrying out weighted fusion on the path centrality of each feature point and the number of times that the feature point is influenced to obtain a network influence score of the feature point, carrying out normalization processing on initial weights of all feature points in the quality feature network based on the network influence score, and completing reassignment of the weights so as to generate the updated quality feature network.
  8. 8. The method for detecting the quality of the rhinitis treatment medicine according to claim 1, wherein the quality judgment process of the rhinitis treatment medicine to be detected is carried out based on the quality characteristic network after weight redistribution, and the method comprises the steps of inputting original sample data of the rhinitis treatment medicine to be detected in a latest batch in the updated quality characteristic network to obtain a corresponding latest quality representation graph, multiplying a state value of each characteristic point in the latest quality representation graph by the weight after weight redistribution to obtain a weighted state value, and calculating the sum of weighted state values of a medicine matrix layer, an active ingredient distribution layer and a physical property representation layer respectively to be used as a comprehensive score of each characteristic layer.
  9. 9. The method according to claim 8, wherein the step of outputting the structured quality report comprises presetting a pass range interval of the composite scores of the feature layers, comparing the composite scores of the calculated drug matrix layers, the composite scores of the active ingredient distribution layers and the composite scores of the physical property presentation layers with the corresponding pass range intervals, respectively, and generating a structured report containing a hierarchical decision result, wherein the structured report explicitly records the composite score value of each feature layer, the corresponding pass range interval, a decision of whether the feature layers are pass or not, and when the feature layers are determined to be fail, the weighted state values of the feature points are obviously abnormal.
  10. 10. A drug quality detection system for the treatment of rhinitis comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that the processor, when executing the computer program, carries out the steps of a drug quality detection method for the treatment of rhinitis as claimed in any of claims 1 to 9.

Description

Medicine quality detection method and system for treating rhinitis Technical Field The invention relates to the technical field of medicine quality detection, in particular to a medicine quality detection method and system for treating rhinitis. Background The quality detection of the current rhinitis treatment medicine depends on the independent measurement of indexes such as spectrum, chromatograph, granularity distribution and the like and the comparison with a standard range. The conventional method is to conduct isolated comparison between each detection data and a preset threshold value or conduct simple multivariate statistical treatment on the data. The technology regards matrix composition, active ingredients and physical properties of the drug as separate inspection items, and can only judge whether single parameters are compliant. The prior art schemes are unable to resolve the inherent associations and dynamic interaction mechanisms between different quality attributes. This results in an insufficient identification of complex quality defects. Meanwhile, a quality judgment system based on a fixed rule or a static model lacks adaptability, and when a sample to be tested presents an atypical abnormal mode or a novel defect, the accuracy and reliability of judgment are limited. The prior detection technology is difficult to realize the deep evaluation of the quality state of the medicine from the system level. What is needed is a detection method that quantifies the dynamic correlation between quality attributes and adaptively adjusts the decision logic based on specific sample characteristics to reveal systematic quality risk and improve the depth and intelligence level of quality control. Disclosure of Invention The invention aims to provide a drug quality detection method and a drug quality detection system for treating rhinitis, which are used for solving the problems in the background technology. To achieve the above object, the present invention provides a drug quality detection method for treating rhinitis, the method comprising: The method comprises the steps of establishing a quality characteristic network, wherein the quality characteristic network consists of a plurality of mutually related characteristic layers and is used for representing the multielement quality attribute of rhinitis drugs, and the quality characteristic network comprises a drug matrix layer, an active ingredient distribution layer and a physical property expression layer, wherein each layer consists of a plurality of quality characteristic points; Obtaining original sample data of rhinitis medicine to be detected, mapping the original sample data to corresponding feature points of the quality feature network, and generating an initial quality characterization graph, wherein the original sample data comprises spectrum data, chromatographic data and granularity distribution data; In the quality characteristic network, carrying out quantization calculation on the association strength between characteristic points in the initial quality characteristic diagram to form a dynamic association link; triggering topological structure adjustment of the quality feature network according to the states of the conduction path and the key nodes to generate an updated quality feature network; based on the quality characteristic network after weight redistribution, executing the quality judgment process of the rhinitis medicine to be detected, and outputting a structured quality report. Preferably, the process of establishing the quality characteristic network comprises the steps of extracting reference sample data from a standard rhinitis drug sample library, carrying out cluster analysis on quality attributes in the reference sample data, identifying three core attribute clusters of drug matrixes, active components and physical properties, defining each core attribute cluster as a characteristic layer, configuring independent characteristic coding rules and characteristic point association rules for each characteristic layer, wherein the characteristic coding rules are used for discretizing continuous quality data into state values of characteristic points, and the characteristic point association rules are used for defining potential connection relations of characteristic points in the same characteristic layer and among different characteristic layers. Preferably, the method comprises the steps of obtaining original sample data of a rhinitis drug to be detected, mapping the original sample data to corresponding feature points of a quality feature network to generate an initial quality characterization graph, specifically, invoking feature coding rules corresponding to a drug matrix layer, an active component distribution layer and a physical property expression layer respectively, carrying out parallel processing on the spectrum data, the spectrum data and the particle size distribution data, mapping the processed spectrum data to