Share Email Print

Journal of Biomedical Optics

Discrimination of healthy and osteoarthritic articular cartilages by Fourier transform infrared imaging and partial least squares-discriminant analysis
Author(s): Xue-Xi Zhang; Jian-Hua Yin; Zhi-Hua Mao; Yang Xia
Format Member Price Non-Member Price
PDF $20.00 $25.00

Paper Abstract

Fourier transform infrared imaging (FTIRI) combined with chemometrics algorithm has strong potential to obtain complex chemical information from biology tissues. FTIRI and partial least squares-discriminant analysis (PLS-DA) were used to differentiate healthy and osteoarthritic (OA) cartilages for the first time. A PLS model was built on the calibration matrix of spectra that was randomly selected from the FTIRI spectral datasets of healthy and lesioned cartilage. Leave-one-out cross-validation was performed in the PLS model, and the fitting coefficient between actual and predicted categorical values of the calibration matrix reached 0.95. In the calibration and prediction matrices, the successful identifying percentages of healthy and lesioned cartilage spectra were 100% and 90.24%, respectively. These results demonstrated that FTIRI combined with PLS-DA could provide a promising approach for the categorical identification of healthy and OA cartilage specimens.

Paper Details

Date Published: 9 June 2015
PDF: 3 pages
J. Biomed. Opt. 20(6) 060501 doi: 10.1117/1.JBO.20.6.060501
Published in: Journal of Biomedical Optics Volume 20, Issue 6
Show Author Affiliations
Xue-Xi Zhang, Nanjing Univ. of Aeronautics and Astronautics (China)
Jian-Hua Yin, Nanjing Univ. of Aeronautics and Astronautics (China)
Zhi-Hua Mao, Nanjing Univ. of Aeronautics and Astronautics (China)
Yang Xia, Oakland Univ. (United States)

© SPIE. Terms of Use
Back to Top