Comparison of the performance of linear multivariate analysis methods for normal and dyplasia tissues differentiation using autofluorescence spectroscopy

Chia Chu Shou, Tzu-Chien Hsiao, Jen K. Lin, Chih Yu Wang, Kenny Chiang Huihua*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

21 Scopus citations

Abstract

We compared the performance of three widely used linear multivariate methods for autofluorescence spectroscopic tissues differentiation. Principal component analysis (PCA), partial least squares (PLS), and multivariate linear regression (MVLR) were compared for differentiating at normal, tubular adenoma/epithelial dysplasia and cancer in colorectal and oral tissues. The methods' performances were evaluated by cross-validation analysis. The group-averaged predictive diagnostic accuracies were 85% (PCA), 90% (PLS), and 89% (MVLR) for colorectal tissues; 89% (PCA), 90% (PLS), and 90% (MVLR) for oral tissues. This study found that both PLS and MVLR achieved higher diagnostic results than did PCA.

Original languageEnglish
Pages (from-to)2265-2273
Number of pages9
JournalIEEE Transactions on Biomedical Engineering
Volume53
Issue number11
DOIs
StatePublished - 1 Nov 2006

Keywords

  • Colorectal tissue
  • Light-induced autofluorescence
  • Multivariate linear regression
  • Oral tissue
  • Partial least squares
  • Principal component analysis

Fingerprint Dive into the research topics of 'Comparison of the performance of linear multivariate analysis methods for normal and dyplasia tissues differentiation using autofluorescence spectroscopy'. Together they form a unique fingerprint.

Cite this