CN-122022549-A - Drug quality detection data processing and analyzing system and method

CN122022549ACN 122022549 ACN122022549 ACN 122022549ACN-122022549-A

Abstract

The application provides a drug quality detection data processing analysis system and method, which are characterized in that time series alignment processing is carried out on original multi-source detection data, detection data fragments reflecting quality attributes of a target drug are further extracted, a characteristic evolution sequence of the quality of the target drug among different batches is generated according to the detection data fragments, a plurality of difference sensitive characteristics representing the stability of the target drug batch are screened out according to the characteristic evolution sequence, all the difference sensitive characteristics are converted into batch-to-batch difference quantities of the target drug in a batch quality inspection process, quality data dispersion of different sample points in the same batch in the aligned multi-source detection data is used for generating uniformity indexes of the quality of the target drug batch, and quality grade analysis is carried out on the target drug of the current batch based on the batch-to-batch difference quantities and the uniformity indexes, so that analysis results are obtained. By adopting the scheme, the comprehensive analysis of the medicine quality batch difference can be realized under the batch quality inspection scene, so that the analysis error in the medicine quality monitoring process is reduced.

Inventors

HU XING
TIAN JIE

Assignees

重庆市食品药品检验检测研究院
重庆市渝北区人民医院

Dates

Publication Date: 20260512
Application Date: 20251231

Claims (10)

1. The medicine quality detection data processing and analyzing method is characterized by comprising the following steps: acquiring original multi-source detection data of a target medicine in a batch quality inspection process; performing time sequence alignment processing on the original multi-source detection data, extracting detection data fragments reflecting quality attributes of the target medicines from the aligned multi-source detection data, and generating characteristic evolution sequences of the quality of the target medicines among different batches according to the detection data fragments; screening out a plurality of difference sensitive features representing the stability of the target medicine batch according to the feature evolution sequence, and converting all the difference sensitive features into inter-batch difference of the target medicine in the batch quality inspection process; generating a uniformity index of quality in the target medicine batch by using quality data dispersion of different sample points in the same batch in the aligned multi-source detection data; And carrying out quality grade analysis on the target medicines in the current batch based on the difference between batches and the uniformity index of the quality in the batch to obtain an analysis result.
2. The method of claim 1, wherein extracting test data segments reflecting quality attributes of the target drug from the aligned multi-source test data comprises: determining a gradient sequence reflecting the change rate of the medicine quality in the aligned multi-source detection data; and extracting detection data fragments reflecting the quality attribute of the target medicine based on the gradient sequence.
3. The method according to claim 1, wherein generating a characteristic evolution sequence of quality of the target drug between different batches from the detection data segment comprises: Extracting multidimensional quality feature vectors from the detected data segments; carrying out standardization processing on the multidimensional quality feature vector; and generating a characteristic evolution sequence of the quality of the target medicine among different batches based on the result obtained by the standardization processing.
4. The method of claim 1, wherein screening out a plurality of differentially sensitive features that characterize stability of a target pharmaceutical lot based on the sequence of feature evolution comprises: Determining the batch-to-batch variability of each feature dimension based on the feature evolution sequence; And screening out a plurality of difference sensitive characteristics representing the stability of the target medicine batch according to the difference degree among all batches.
5. The method of claim 1, wherein converting all of the variance-sensitive features into inter-lot variance measures of the target drug during the lot quality inspection process comprises: determining the feature distance between the current batch and the reference batch of each difference sensitive feature; and determining the batch-to-batch difference of the target medicine in the batch quality inspection process based on all the characteristic distances.
6. The method of claim 1, wherein generating a uniformity index for quality within the target drug lot from quality data dispersion of different sample points within the same lot in the aligned multi-source test data specifically comprises: determining quality data dispersion of different sample points in the same batch in the aligned multi-source detection data; And determining the uniformity index of the quality in the target medicine batch according to the quality data dispersion.
7. The method of claim 1, wherein performing quality level analysis on the target drug of the current lot based on the inter-lot variance and the uniformity index of the intra-lot quality, the analysis result specifically comprising: Determining a decision matrix based on the inter-batch variation and a uniformity index of the intra-batch quality; determining a quality grade according to the position of the quality detection data of the target medicines in the current batch in the judgment matrix; And generating an analysis result based on the quality grade.
8. A pharmaceutical quality inspection data processing analysis system, comprising: The acquisition module is used for acquiring original multi-source detection data of the target medicine in the batch quality inspection process; The processing module is used for carrying out time sequence alignment processing on the original multi-source detection data, extracting detection data fragments reflecting the quality attribute of the target medicine from the aligned multi-source detection data, and generating a characteristic evolution sequence of the quality of the target medicine among different batches according to the detection data fragments; the processing module is further used for screening out a plurality of difference sensitive features representing the stability of the target medicine batch according to the feature evolution sequence, and converting all the difference sensitive features into batch-to-batch difference of the target medicine in the batch quality inspection process; The processing module is further used for generating a uniformity index of the quality in the target medicine batch according to the quality data dispersion of different sample points in the same batch in the aligned multi-source detection data; And the execution module is used for carrying out quality grade analysis on the target medicines in the current batch based on the inter-batch difference and the uniformity index of the quality in the batch to obtain an analysis result.
9. A computer device, characterized in that the computer device comprises a memory for storing a computer program and a processor for calling and running the computer program from the memory, so that the computer device performs the pharmaceutical quality detection data processing analysis method according to any one of claims 1 to 7.
10. A computer-readable storage medium having instructions or code stored therein which, when executed on a computer, cause the computer to perform the pharmaceutical quality inspection data processing analysis method of any one of claims 1 to 7.

Description

Drug quality detection data processing and analyzing system and method Technical Field The application relates to the technical field of data processing and analysis, in particular to a system and a method for processing and analyzing medicine quality detection data. Background The medicine quality detection is a systematic judgment process for internal structure change, component fluctuation condition and batch consistency of the medicine through physicochemical analysis, impurity detection, stability test and multidimensional quality index evaluation. The method has the core aim of ensuring stable quality performance of the medicines in production, storage, transportation and use periods and providing a reliable data base for quality supervision, production consistency control and post-marketing evaluation. With the increasing number of drug types and production lots, drug quality detection data exhibit multi-source, time-sequential and dynamic evolution characteristics, so that how to efficiently process and analyze the data becomes a key link in a quality management system. In the prior art, the processing and analysis of the drug quality detection data mostly takes static data statistics analysis as a core, index calculation is usually carried out only for detection data of a single time point or a fixed segment, dynamic evolution rules of drug quality attributes in the whole production and detection process are ignored, integrated analysis of multi-dimensional data is difficult to realize, in addition, the traditional method depends on experience threshold values or single statistics index during feature screening, key features sensitive to drug batch stability are difficult to accurately identify, the method has the advantages that deviation exists in batch-to-batch difference evaluation, the existing method cannot comprehensively capture dynamic change characteristics of medicine quality, the multi-source data integration effect is poor, evaluation accuracy of batch-to-batch difference and uniformity in batches is insufficient, medicine quality conditions are difficult to truly reflect, the scientificity of production quality control decision is further influenced, risks of misjudgment of qualified batches and missed judgment of unqualified batches are increased, and hidden danger is brought to medicine quality safety; therefore, how to realize comprehensive analysis of drug quality batch differences in a batch quality inspection scene so as to reduce analysis errors in the drug quality monitoring process becomes a difficult problem in the industry. Disclosure of Invention The application provides a drug quality detection data processing analysis system and a drug quality detection data processing analysis method, which can realize comprehensive analysis of drug quality batch differences under a batch quality inspection scene so as to reduce analysis errors in a drug quality monitoring process. In a first aspect, the present application provides a method for processing and analyzing drug quality detection data, comprising the steps of: acquiring original multi-source detection data of a target medicine in a batch quality inspection process; performing time sequence alignment processing on the original multi-source detection data, extracting detection data fragments reflecting quality attributes of the target medicines from the aligned multi-source detection data, and generating characteristic evolution sequences of the quality of the target medicines among different batches according to the detection data fragments; screening out a plurality of difference sensitive features representing the stability of the target medicine batch according to the feature evolution sequence, and converting all the difference sensitive features into inter-batch difference of the target medicine in the batch quality inspection process; generating a uniformity index of quality in the target medicine batch by using quality data dispersion of different sample points in the same batch in the aligned multi-source detection data; And carrying out quality grade analysis on the target medicines in the current batch based on the difference between batches and the uniformity index of the quality in the batch to obtain an analysis result. With reference to the first aspect, in one possible implementation manner, extracting a detection data segment reflecting a quality attribute of the target drug from the aligned multi-source detection data specifically includes: determining a gradient sequence reflecting the change rate of the medicine quality in the aligned multi-source detection data; and extracting detection data fragments reflecting the quality attribute of the target medicine based on the gradient sequence. With reference to the first aspect, in one possible implementation manner, generating a feature evolution sequence of quality of the target drug between different batches according to the detection data segment specifical