指纹(计算)
偏最小二乘回归
化学计量学
主成分分析
线性判别分析
人工智能
模式识别(心理学)
高效液相色谱法
数学
化学
色谱法
计算机科学
统计
作者
Miao Liu,Wanzhen Xiao,Hong Zhang,Guoxiang Sun
标识
DOI:10.1016/j.saa.2022.121968
摘要
The study aimed to provide a reliable and feasible strategy for the comprehensive quality control of medicine food homology materials (MFHM). The high performance liquid chromatography (HPLC) fingerprints and Fourier transform mid-infrared (FT-MIR) quantized fingerprints were successfully developed to comprehensively evaluate overall quality of Citri Reticulata Pericarpium (CRP) by applying comprehensive quantified fingerprint method (CQFM). All samples were well distinguished and divided into 5 grades. In addition, through principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA), the identification ability of HPLC fingerprints and FT-MIR fingerprints on CRP with different storage years was discussed. The results showed that HPLC fingerprints combined with PCA had good discrimination ability, and the PLS-DA model established by the preprocessed FT-MIR fingerprint data could accurately distinguish and predict the storage period of CRP. Finally, based on 1, 1-diphenyl-2-picrylhydrazyl radical (DPPH•) scavenging assay, combined with bivariate correlation analysis, the fingerprint-activity relationship of offline antioxidant activity of CRP samples with the fingerprints peak were studied. In general, the comprehensive strategies provide a reliable and scientific reference scheme for the quality control of MFHM in the future.
科研通智能强力驱动
Strongly Powered by AbleSci AI