Gregor Peltri | PubMed Health

PubMed2003

Increased predictive value of parameters by fuzzy logic-based multiparameter analysis.

BACKGROUND: A recent study on postoperative effusions and edema was used to demonstrate the potential of fuzzy techniques in multiparameter data analysis. In this study, more than 50 parameters of 75 patients were collected and examined for correlations between some of the parameters and the later development of complications. METHODS: We employed a rule-based fuzzy-logic system in order to combine the diagnostic values of single parameters. The advantage of fuzzy sets is that they substitute sharp cut-off values with a smooth transition from one property to another. Therefore, there is no decision of "either-or" but rather a graded assessment of "more or less", which is often more suitable for a problem. RESULTS: The fuzzy combination of parameters led to a large increase of sensitivity and specificity when compared with the best single parameter. This increase was achieved by taking a close look at the parameters. A newly created parameter, relative weight, turned out to be very powerful. CONCLUSIONS: Fuzzy techniques can increase the discriminating power of classical statistical tools. In addition, results obtained by fuzzy analysis are highly interpretable. A combination of the CLASSIF1 algorithm for the identification of the most relevant parameters, followed by fuzzy analysis, represents a powerful tool for the handling of large amounts of multiparameter data.

Adolescent↗

PubMedDate not supplied

Fuzzy logic-based tumor marker profiles including a new marker tumor M2-PK improved sensitivity to the detection of progression in lung cancer patients.

In lung cancer patients tumor markers are used for disease monitoring. The goal of this study was to improve diagnostic efficiency in the detection of tumor progression in lung cancer patients by using fuzzy logic modeling in combination with a tumor marker panel (Tumor M2-PK, CYFRA 21-1, CEA, NSE and SCC). Thirty-three small cell lung cancers (SCLC) and 69 consecutive inoperable patients (40 squamous and 29 adenocarcinomas) were included in a prospective study. The changes of blood levels of tumor markers as well as their analysis by fuzzy logic modelling were compared to the clinical evaluation of response vs. non-response to therapy. Clinical monitoring was evaluated according to the standard criteria of the WHO. Tumor M2-PK was measured in plasma with an ELISA (ScheBo Biotech, Germany) and all other markers in sera (Roche, Germany). At a 90% specificity, the respective best single marker found the following fraction of all patients who had tumor progression clinically detected: in SCLC with NSE 52%, in adenocarcinoma with CYFRA 21-1 89% and in squamous carcinoma with SCC 65%. A fuzzy logic rule-based system employing a tumor marker panel increased the sensitivity in small cell carcinomas to 73% with the marker combination NSE/CEA and to 63% with the marker combination NSE/Tumor M2-PK, respectively. In squamous carcinomas an improvement of sensitivity is also observed using the marker combination of SCC/Tumor M2-PK (Sensitivity: 81%) or SCC/CEA (Sensitivity: 71%). By using the fuzzy logic method and the marker combination CYFRA 21-1/CEA as well as CYFRA 21-1/Tumor M2-PK, the detection of lung cancer progression was possible in all adenocarcinomas. With the fuzzy logic method and a tumor marker panel (including the new marker Tumor M2-PK), a useful diagnostic tool for the detection of progression in lung cancer patients is available.

Adenocarcinoma↗