The increasing use of Electroencephalography (EEG) in the field of pediatric neurology allows more accurate and precise diagnosis of several cerebral pathologies, mainly in Neonatal Intensive Care Units (NICUs), where it represents the gold-standard for the diagnosis of neonatal epileptic seizures. However, EEG interpretation is time consuming and requires highly specialized staff. For this reason, in the last years there was a growing interest in the development of systems for automatic and fast detection of neonatal epileptic seizures. To this aim, we propose here hybrid systems that combines techniques related to the Stationary Wavelet Transform (SWT) as a support to deep-learning algorithms such as Convolutional Neural Networks and Fully Convolutional Networks. The proposed methods are validated on a public dataset of NICUs seizures recorded at the Helsinki University Hospital. Results are encouraging both in terms of Area Under the receiveroperating Curve, AUC (81%), Good Detection Rate, GDR (77%) and False Detection per hour, FD/h (1.6). Actually, the SWT step increases the performance of the proposed methods of about 5% for the AUC as compared to considering the raw EEG time-series only. These results, though preliminary, represent a significant step forward in solving the problem of neonatal seizure detection.

No abstract

The complex physiological dynamics of neonatal seizures make their detection challenging. A timely diagnosis and treatment, especially in intensive care units, are essential for a better prognosis and the mitigation of possible adverse effects on the newborn’s neurodevelopment. In the literature, several electroencephalographic (EEG) studies have been proposed for a parametric characterization of seizures or their detection by artificial intelligence techniques. At the same time, other sources than EEG, such as electrocardiography, have been investigated to evaluate the possible impact of neonatal seizures on the cardio-regulatory system. Heart rate variability (HRV) analysis is attracting great interest as a valuable tool in newborns applications, especially where EEG technologies are not easily available. This study investigated whether multiscale HRV entropy indexes could detect abnormal heart rate dynamics in newborns with seizures, especially during ictal events. Furthermore, entropy measures were analyzed to discriminate between newborns with seizures and seizure-free ones. A cohort of 52 patients (33 with seizures) from the Helsinki University Hospital public dataset has been evaluated. Multiscale sample and fuzzy entropy showed significant differences between the two groups (p-value < 0.05, Bonferroni multiple-comparison post hoc correction). Moreover, interictal activity showed significant differences between seizure and seizure-free patients (Mann-Whitney Test: p-value < 0.05). Therefore, our findings suggest that HRV multiscale entropy analysis could be a valuable pre-screening tool for the timely detection of seizure events in newborns.

No abstract