基于全血可见吸收光谱的物种鉴别

doi:10.3964/j.issn.1000-0593(2017)11-3510-04

Abstract
References
Related Citation (15)

Download: PDF (1868 KB)
Export: BibTeX | EndNote (RIS)

Abstract Blood and blood products which are special products should be controlled strictly in the process of circulation. The traditional methods to identify the species of the blood use antihuman hemoglobin(Hb) antibodies which recognize and bind to human Hb, or use high-performance liquid chromatography(HPLC) and mass spectrometry (MS). During the identification, the samples may exposure to the external environment. To avoid the pollution of the samples and the environment, a non-connect technique for species of blood identification is needed. The experiment tests 24 human and bovine whole blood samples, including 10 samples for each and 4 unknown samples. The PET tube which contains the blood sample sets in the integrating sphere, which is illuminated by the supercontinuum laser. Each sample takes 5 times of the absorb spectrogram as average in range of 420~740 nm. A PCA method is used to reduce the dimension of original spectral data. Finally, the reduced data are fitted by two-dimensional normal distribution. The species of unknown samples can be identified by using the probability distribution in confidence space or the distance to the known classification center. In conclusion, the visible absorb spectrum of whole blood can be well identified by using the combination of PCA and two-dimensional normal distribution.

Key words：Absorb spectrum; Visible light; PCA; Normal distribution

Received: 2016-09-26 Accepted: 2017-02-10

ZTFLH:

O675.3

Corresponding Authors: WAN Xiong E-mail: wanxiong@mail.sitp.ac.cn

Cite this article:

URL:

https://www.gpxygpfx.com/EN/10.3964/j.issn.1000-0593(2017)11-3510-04 OR https://www.gpxygpfx.com/EN/Y2017/V37/I11/3510