Add to Calendar
When:
27. November 2020 @ 14:00 – 15:00
2020-11-27T14:00:00+01:00
2020-11-27T15:00:00+01:00
Where:
Zoom
Cost:
Free
Contact:
Dr. Max-Philipp Stenner
onLINe Seminar
Alain de Cheveigne (PI Paris)
>>Brain Data Analysis and Decoding<<
Alain has developed many highly effective signal processing techniques, in particular for denoising electrophysiological data. Some of his blind source separation techniques are already in use at the VNeu and SPL, and we thought that many others at LIN might profit from his techniques – see, for example, his recent Neuron paper on pitfalls of, and alternatives to, spectral filtering.
Abstract
Techniques to measure brain activity (EEG, MEG, LFP, ECoG, etc.) all share a similar set of issues: high levels of noise, mixing between sources and observations, and limited dimensionality relative to the billions of sources active within the brain. Finding a meaningful response sometimes feels like searching for a needle in a haystack. I will discuss approaches to address these issues. Source to sensor mixing is highly linear, and thus linear analysis methods are prominent, aiming to attempt to reverse the mixing or factor out major sources of noise and artifact. A wide range of techniques is available, including PCA, ICA, joint decorrelation, CCA, and others. I will discuss my favorite methods, explain how they work and what they can do, and look at how we might overcome present-day limitations, to dig deeper into the haystack in search of the needle.
Techniques to measure brain activity (EEG, MEG, LFP, ECoG, etc.) all share a similar set of issues: high levels of noise, mixing between sources and observations, and limited dimensionality relative to the billions of sources active within the brain. Finding a meaningful response sometimes feels like searching for a needle in a haystack. I will discuss approaches to address these issues. Source to sensor mixing is highly linear, and thus linear analysis methods are prominent, aiming to attempt to reverse the mixing or factor out major sources of noise and artifact. A wide range of techniques is available, including PCA, ICA, joint decorrelation, CCA, and others. I will discuss my favorite methods, explain how they work and what they can do, and look at how we might overcome present-day limitations, to dig deeper into the haystack in search of the needle.