5 November 2015 Classification and identification of amino acids based on THz spectroscopy
Author Affiliations +
Proceedings Volume 9795, Selected Papers of the Photoelectronic Technology Committee Conferences held June–July 2015; 97953I (2015) https://doi.org/10.1117/12.2214279
Event: Selected Proceedings of the Photoelectronic Technology Committee Conferences held June-July 2015, 2015, Hefei, Suzhou, and Harbin, China
Abstract
Amino acids are important nutrient substances for life, and many of them have several isomerides, while only L-type amino acids can be absorbed by body as nutrients. So it is certain worth to accurately classify and identify amino acids. In this paper, terahertz time-domain spectroscopy (THz-TDS) was used to detect isomers of various amino acids to obtain their absorption spectra, and their spectral characteristics were analyzed and compared. Results show that not all isomerides of amino acids have unique spectral characteristics, causing the difficulty of classification and identification. To solve this problem, partial least squares discriminant analysis (PLS-DA), firstly, was performed on extracting principal component of THz spectroscopy and classifying amino acids. Moreover, variable selection (VS) was employed to optimize spectral interval of feature extraction to improve analysis effect. As a result, the optimal classification model was determined and most samples can be accurately classified. Secondly, for each class of amino acids, PLS-DA combined with VS was also applied to identify isomerides. This work provides a suggestion for material classification and identification with THz spectroscopy.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Ping J. Huang, Ping J. Huang, Ye H. Ma, Ye H. Ma, Xian Li, Xian Li, Di B. Hou, Di B. Hou, Jin H. Cai, Jin H. Cai, Guang X. Zhang, Guang X. Zhang, } "Classification and identification of amino acids based on THz spectroscopy", Proc. SPIE 9795, Selected Papers of the Photoelectronic Technology Committee Conferences held June–July 2015, 97953I (5 November 2015); doi: 10.1117/12.2214279; https://doi.org/10.1117/12.2214279
PROCEEDINGS
9 PAGES


SHARE
Back to Top