Classification of human gliomas by infrared imaging spectroscopy and chemometric image processing

As a molecular probe of tissue composition, infrared spectroscopic imaging can potentially serve as an adjunct to histopathology in detecting and diagnosing disease. This study demonstrates that human gliomas are distinguishable from control tissue on the basis of IR image used in combination with c...

Full description

Saved in:
Bibliographic Details
Published in:Vibrational spectroscopy 2005-07, Vol.38 (1), p.143-149
Main Authors: Beleites, C., Steiner, G., Sowa, M.G., Baumgartner, R., Sobottka, S., Schackert, G., Salzer, R.
Format: Article
Language:English
Subjects:
Brain tumors
Chemometric imaging
Classification
Combining classifiers
Grading
Infrared spectroscopic imaging
Small sample size
Validation
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:As a molecular probe of tissue composition, infrared spectroscopic imaging can potentially serve as an adjunct to histopathology in detecting and diagnosing disease. This study demonstrates that human gliomas are distinguishable from control tissue on the basis of IR image used in combination with chemometric imaging processing. Using an iterative two-step algorithm – comprised of a linear discriminant analysis guided genetic optimal spectral region selection – tissue types can be discriminated from one another thus providing insight into the malignancy grade of the tissue. A series of classification models were built using a k-fold cross validation scheme and the classification predictions from the various models were combined to provide an aggregated prediction. The validation of the aggregated model reveals an improvement in the classification success rate to 64%.
ISSN:0924-2031
1873-3697
DOI:10.1016/j.vibspec.2005.02.020