Paper
8 February 2010 Functional data analysis for endoscopic spectral measurements
Vladimir Bochko, Maarit Luukkainen-Soilu, Petri Välisuo, Olli A. T. Peltoniemi, Per H. Rosenberg, Antti K. Sukura, Pernilla Syrjä, Thomas Spillmann, Jarmo Alander
Author Affiliations +
Abstract
We use endoscopy for VIS-NIR spectral analysis of the porcine gastrointestinal wall. A flexible endoscope was implemented with a fiber optic probe for measuring spectra. The spectral data were measured by a spectrophotometer (Ocean Optics HR4000). The presented measurements are the spectra from three locations of the gastrointestinal wall of a healthy pig (gastric corpus and antrum, duodenum) taken in a retrograde order. We tested several features obtained by functional data analysis. The features were linearly combined using principal component analysis (PCA) and partial least squares (PLS). Finally, the support vector classifier (SVC) and PLS were used to classify spectra from corpus, antrum and duodenum. The experiments confirm the feasibility of our methods for classification of spectra of pig intestines.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Vladimir Bochko, Maarit Luukkainen-Soilu, Petri Välisuo, Olli A. T. Peltoniemi, Per H. Rosenberg, Antti K. Sukura, Pernilla Syrjä, Thomas Spillmann, and Jarmo Alander "Functional data analysis for endoscopic spectral measurements", Proc. SPIE 7547, Saratov Fall Meeting 2009: International School for Junior Scientists and Students on Optics, Laser Physics, and Biophotonics, 754702 (8 February 2010); https://doi.org/10.1117/12.852750
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Principal component analysis

Data analysis

Endoscopy

Medicine

Spectrophotometry

Endoscopes

Intestine

Back to Top