Time progression of a parametric impulse response function estimate from intra-partum cardiotocography for normal and hypoxic fetuses

We compared the time evolution of a parametric system identification model of cardiotocography (CTG) during the last three hours of labour and delivery. The models are the impulse response functions (IRFs) of the system relating the uterine pressure (UP) and fetal heart rate (FHR). We used surrogate data to assess the likelihood that the model successfully identified system dynamics. We then investigated the ability of the models to indicate fetal distress by comparing normal and hypoxic cases. For pathological and normal fetuses, model output standard deviation σ
ŷ
, first IRF minimum time Tmin and variance accounted for (VAF) were discriminating in the first two hours. In the final hour however, only T
min
remained discriminating. For intermediate and normal cases, σ
ŷ
was always discriminating while T
min
discriminated in the final hour only. These results suggest that detecting some pathological and intermediate cases before injury occurs may be possible; this would be very useful clinically.