Fetal Heart Rate Feature Extraction from Cardiotocographic Recordings through Autoregressive Model's Power Spectral-and Pole-based Analysis

Alfredo Illanes (1) , Michel Haritopoulos (2)

1 Universidad Austral de Chile
2 PRISME - Laboratoire Pluridisciplinaire de Recherche en Ingénierie des Systèmes, Mécanique et Energétique

— The main objective of this work is to perform an autoregressive model (AR)-based power spectral analysis of the fetal heart rate (FHR) signal for the extraction of significant features for fetal welfare assessment. A group of features is directly computed from the AR-based spectrum while another group is computed from the poles representation. The presented method is applied to real cardiotocographic (CTG) signals and for different frequency bands, and the obtained results are very promising as they exhibit direct correlations between the extracted features and the fetal welfare in terms of umbilical pH.

Domaines: Traitement du signal et de l'image [eess.SP]

Alfredo Illanes, Michel Haritopoulos. Fetal Heart Rate Feature Extraction from Cardiotocographic Recordings through Autoregressive Model's Power Spectral-and Pole-based Analysis. 37th annual international conference of the IEEE Engineering in Medicine and Biology Society - EMBC 2015, Aug 2015, Milan, Italy. ⟨hal-01203022⟩