CLASSIFICATION OF FETAL HEART-RATE SEQUENCES BASED ON FRACTAL FEATURES

Visual inspection of foetal heart rate (FHR) sequences is an important means of foetal well-being evaluation. The application of fractal fea tures for classifying physiologically relevant FHR sequence patterns i s reported. The use of fractal features is motivated by the difficulti es exhibited by traditional classification schemes to discriminate som e classes of FHR sequence and by the recognition that this type of sig nal exhibits features on different scales of observation, just as frac tal signals do. To characterise the signals by fractal features, two a pproaches are taken. The first models the FHR sequences as temporal fr actals. The second uses techniques from the chaos-theory field and aim s to model the attractor based on FHR sequences. The fractal features determined by both approaches are used to design a Bayesian classifica tion scheme. Classification results for three classes are presented; t hey are quite satisfactory and illustrate the importance of this type of methodology.