EURASIP Journal on Advances in Signal Processing (Jan 2010)
Time-Frequency Characterization of Cerebral Hemodynamics of Migraine Sufferers as Assessed by NIRS Signals
Abstract
Near-infrared spectroscopy (NIRS) is a noninvasive system for the real-time monitoring of the concentration of oxygenated (O2Hb) and reduced (HHb) hemoglobin in the brain cortex. O2Hb and HHb concentrations vary in response to cerebral autoregulation. Sixty-eight women (14 migraineurs without aura, 49 migraineurs with aura, and 5 controls) performed breath-holding and hyperventilation during NIRS recordings. Signals were processed using the Choi-Williams time-frequency transform in order to measure the power variation of the very-low frequencies (VLF: 20–40 mHz) and of the low frequencies (LF: 40–140 mHz). Results showed that migraineurs without aura present different LF and VLF power levels than controls and migraineurs with aura. The accurate power measurement of the time-frequency analysis allowed for the discrimination of the subjects' hemodynamic patterns. The time-frequency analysis of NIRS signals can be used in clinical practice to assess cerebral hemodynamics.