Multiscale and wavelet based signal processing for electrophysiological and optical brain imaging

点击翻译按钮获取中文摘要

Medical imaging, and functional neuroimaging in particular, is one of the most dynamic of the scientific and technological fields in biomedical. Bioelectrical measurements acquired on the scalp of the head are vitally important in terms of responding to the challenges of neurological illness, as well as for designing technologies capable of decoding neural activity, control systems for the disabled, for example. Electrophysiology, the study of the electrical properties of biological organisms, has developed considerably since the first measurements of electrical potential were recorded by Richard Caton (1875), and some very useful tools for measuring neural activity have been devised as a result. Over the last thirty years, one of these technologies, magnetoencephalography (measurement of the magnetic field created by neural activity), has been combined with another tool, electroencephalography, to provide the 'bioelectromagmetic signature' of the activity of the brain. A third technology, based on the diffusion and absorption of infrared light by perfused brain tissues, makes it possible to measure the hemodynamic variations caused by neural activity. Such measurements are essential in evaluating a pathology in a clinical setting, like epilepsy, for example, or in the neurosciences and cognitive sciences, in which the brain functions are studied in particular situations, like during sleep or when the brain is engaged in tasks involving perception or memory, for example. This research program focuses on the digital tools that make it possible to process optical and bioelectomagnetic measurements on a computer, in order to visualize the regions of the cortex involved in the complex and dynamic networks that constitute brain activity. We will also study the intracranial electrophysiological measurements recorded in epileptic patients evaluated within the framework of a surgery, by studying the bursts of rapid oscillations that occur, and seem, some of them at least, to be related to the pathology. To characterize these signals from the point of view of their spectral complexity at the head surface is one of the challenges of this research program, which is aimed at developing new methods of analyzing the 'working human brain'.