Hilbert-Huang transform: Preliminary studies in epilepsy and cardiac arrhythmias

This work describes a preliminary study based on a time-frequency analysis tool, the Hilbert-Huang transform, which can be applied to the analysis of signals deemed, to some degree, nonstationary and nonlinear. Through implementation in a software programming environment and testing on a sample of patient data sets, we demonstrate the potential of this technique. Its ability to naturally decompose electrical signals of both the heart (ECG) and brain (invasive EEG) into clear frequency and amplitude variations are shown. This ability is reflected in provisional results in the detection of epileptic seizures and cardiac arrhythmias.