We propose new classifiers based on fuuy inference systems (FISs) for the Fetal Heart Rate (FHR) signal analysis. They include standard cardiotocographic (CTG) parameters together with a set of ffequency domain and nonlinear indices. The goal is the identification of two very common fetal pathological conditions: Intra-Uterine Growth Retardation (IUGR) and Diabetes type I. The FHR signals obtained from 104 CTG recordings were analyzed (75 Normal, 11 IUGR and 18 Diabetic). Fuzzy classifiers combine the set of 10 input data into the 3-output set (Normal, IUGR, Maternal Diabetes) by fuzzy rules. Results show FlSs predict normal and pathological fetal states even with 100% of correct classifications. Their performance however is always higher than 80% in the whole population, depending on the rule number. This approach can strongly help the automatic CTG signal analysis improving the early discrimination among normal and pathological fetal conditions.
Classification of fetal pathologies through fuzzy inference systems based on a multiparametric analysis of fetal heart rate / M.G. Signorini, A. de Angelis, G. Magenes, R. Sassi, D. Arduini, S. Cerutti - In: Computers in Cardiology 2000[s.l] : IEEE Press, 2000. - ISBN 0780365577. - pp. 435-438 (( Intervento presentato al 27. convegno Computers in Cardiology tenutosi a Cambridge nel 2000.
Classification of fetal pathologies through fuzzy inference systems based on a multiparametric analysis of fetal heart rate
R. Sassi;
2000
Abstract
We propose new classifiers based on fuuy inference systems (FISs) for the Fetal Heart Rate (FHR) signal analysis. They include standard cardiotocographic (CTG) parameters together with a set of ffequency domain and nonlinear indices. The goal is the identification of two very common fetal pathological conditions: Intra-Uterine Growth Retardation (IUGR) and Diabetes type I. The FHR signals obtained from 104 CTG recordings were analyzed (75 Normal, 11 IUGR and 18 Diabetic). Fuzzy classifiers combine the set of 10 input data into the 3-output set (Normal, IUGR, Maternal Diabetes) by fuzzy rules. Results show FlSs predict normal and pathological fetal states even with 100% of correct classifications. Their performance however is always higher than 80% in the whole population, depending on the rule number. This approach can strongly help the automatic CTG signal analysis improving the early discrimination among normal and pathological fetal conditions.File | Dimensione | Formato | |
---|---|---|---|
00898550.pdf
accesso riservato
Tipologia:
Publisher's version/PDF
Dimensione
345.79 kB
Formato
Adobe PDF
|
345.79 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.