Detecting in-born metabolic disease in newborns by pattern classification of NMR spectra

Schliep, Alexander and Castiglione, Filippo (1999) Detecting in-born metabolic disease in newborns by pattern classification of NMR spectra.
Technical Report p.

Abstract

We classify 1H-nuclear magnetic resonance spectra of urine samples obtained from new-born babies with respect to inborn metabolic diseases. Due to limitations of the training data set, only the class of spectra from healthy subjects can be modeled. We provide two distinct decision functions for membership in this class, which both result in high recognition rates of the disease.


Actions:
Full text not available from this repository.
Export as:
Editorial actions: View Item View Item (Login required)
Deposit Information:
ZAIK Number: zaik1999-364
Depositing User: Alexander Schliep
Date Deposited: 24 Apr 2001 00:00
Last Modified: 19 Aug 2011 11:34
URI: http://e-archive.informatik.uni-koeln.de/id/eprint/364