|
Large-Scale of MEG/EEG in Cognitive Neurology. Challenges and Opportunities| old_uid | 16536 |
|---|
| title | Large-Scale of MEG/EEG in Cognitive Neurology. Challenges and Opportunities |
|---|
| start_date | 2018/10/15 |
|---|
| schedule | 11h |
|---|
| online | no |
|---|
| location_info | Bât. IDEE, Amphithéâte Margaux Hemingway |
|---|
| details | Cycle des conférences 2018 |
|---|
| summary | Magnetoencephalography (MEG) and electroencephalography (EEG) can access population- level neuronal dynamics across vast temporal scales. With the advent of machine learning, M/EEG has become a promising resource for predictive modeling of cognitive states and neuro-psychiatric conditions due to the intimate link between brain rhythms and organized behavior. While MEG can recover spatial patterns at higher signal-to-noise ratio (SNR) and enjoys a more selective cortical resolution, the high availability and portability of EEG makes it a promising candidate for "big data" approaches to cognition. Processing M/EEG data, however, is inherently challenging due to its high-dimensional nature and low SNR of brain- related M/EEG occurrences. Unfortunately, to date, various steps of M/EEG processing pipelines have not yet been fully automated, which commonly gives rise to parameter "hacking" and manual labor. This not only incurs high costs in human processing time but also makes research findings less reproducible. The fact that M/EEG is processing-intense and its data commonly undergo a longer chain of transformations before visualization or statistical analysis does not simplify the situation. Throughout this lecture I will share insights on how machine learning can help improve M/EEG analysis from data cleaning over source localization to predictive modeling of neurological conditions. I will present two recently developed algorithms which automatize parameter estimation for artifact rejection [1] and covariance estimation [2] for source localization. Subsequently, I will present our latest EEG study on robust machine learning of diagnosis in severely brain injured patients [3], and our new MEG study on neuronal recycling of the early blind’s visual cortex for semantic processing. I will conclude by throwing a glance at the open source MNE-Software for processing M/EEG data [5] (http://mne-tools.github.io/stable/index.html), our ongoing community efforts targeting at programmatic access to large-scale data resources, such as the Human Connectome Project (http://mne-tools.github.io/mne-hcp/) and practical concepts for reproducible group analysis [5] (http://mne-tools.github.io/mne-biomag-group- demo/). |
|---|
| responsibles | CRNL |
|---|
| |
|