SearchPublisher and Institutes
You are here: Home :: Area NEM :: Medical science :: Human medicine
Norman Bitterlich
Tumormarker-Profilauswertung – und aus Messdaten wird Information. Analysis of tumor marker profiles – turning measured data into information
Keywords: Entscheidungshilfesystem, fuzzybasierte Mustererkennung, Sensitivit?tserhöhung, Tumormarker-Profilauswertung
In practical use, the experiences of single marker interpretation influence the analysis of tumor marker profiles. From a mathematical point of view, there are different possibilities to link individual measured data. When comparing such methods of resolution, significant differences in their efficiency are observed. An increase in sensitivity generally results in a loss of specificity. Therefore, the choice of the mathematical process is decisively influenced by subjective expectations.
The goal of the analysis of tumor marker profiles is raising the sensitivity while maintaining a high specificity. Based on the study of frequently used strategies of data evaluation, effects are described from which a universally valid concept of data evaluation can be derived. Special attention is given to the independence of the mathematical processes from specified cut-offs and thereby from diagnostic testing methods. Instead of quantitative measured data, qualitative valuations of decision security are handled. This not only enables a rise in sensitivity but also allows a plausible interpretation of results from the user's view. Thus the analysis of tumor marker profiles seems (fuzzy) logical and opens up new avenues to make far-reaching contributions to the clarification of clinical problems by laboratory analysis.
LaboratoriumsMedizin, Walter de Gruyter
Print ISSN: 0025-8466
Volume: 29, 04/2005
Pages: 108 - 112