Pattern recognition methods in evaluation of the structure of the laboratory data biominerals, antioxidant enzymes, selected biochemical parameters, and pulmonary function of welders

M Stepniewski, E Kolarzyk, P Zagrodzki, K Solarska, A Pietrzycka, Mariusz Kitlinski, M Schlegel-Zawadzka, B Batko

Research output: Contribution to journalArticlepeer-review

Abstract

A wide set of biochemical and spirometrical parameters was applied to a group of 41 welders. The obtained parameters were analyzed by the pattern recognition method: principal-components analysis (PCA) and mutual contribution analysis (MCA). The aim of the investigation was to find out which of 23 parameters had an influence on the pulmonological indices of COPD. It was found that PCA gave a substantial reduction of dimensionality. The most relevant parameters were Fe, Zn, Cu (all elements in blood, not in hair), FEV1, MEF50, bilirubin, TAS, and SOD. Only the relationships of MEF50, Cu, and TAS were statistically valid.
Original languageEnglish
Pages (from-to)39-46
JournalBiological Trace Element Research
Volume93
Issue number1-3
DOIs
Publication statusPublished - 2003

Subject classification (UKÄ)

  • Gynaecology, Obstetrics and Reproductive Medicine

Free keywords

  • Principal-components analysis (PCA)
  • (MCA)
  • mutual contribution analysis
  • copper
  • zinc
  • iron
  • magnesium
  • blood serum
  • hair
  • total
  • welders
  • antioxidant status (TAS)
  • chronic obstructive pulmonary disease (COPD)

Fingerprint

Dive into the research topics of 'Pattern recognition methods in evaluation of the structure of the laboratory data biominerals, antioxidant enzymes, selected biochemical parameters, and pulmonary function of welders'. Together they form a unique fingerprint.

Cite this