|
|
|
| e-Pub |
Previous | 
Home
Section: New Results
From the Mesoscopic to the Macroscopic Scale
Participants: Laurent Bougrain, Axel Hutt, Tamara Tošić, Mariia Fedotenkova, Meysam Hashemi, Cecilia Lindig-Leon, Jimmy, Nex, Sébastien Rimbert.
In collaboration with Stéphanie Fleck (Univ. Lorraine), Nathalie Gayraud (Inria Sophia Antipolis) and Maureen Clerc (Inria Sophia Antipolis)
Level of Consciousness
Participants: Axel Hutt, Meysam Hashemi
Meysam Hashemi defended his thesis about analytical and numerical studies of thalamo-cortical neural population models during general anesthesia. The findings of this thesis provide new insights into the mechanisms responsible for the specific changes in EEG patterns that are observed during propofol-induced sedation. Our results indicate that depending on the mean potential values of the system resting states, an increase or decrease in the thalamo-cortical gain functions results in an increase or decrease in the alpha power, respectively. In contrast, the evolution of the delta power is rather independent of the system resting states; the enhancement of spectral power in delta band results from the increased synaptic or extra-synaptic GABAergic inhibition. Furthermore, we aim to identify the parameters of a thalamo-cortical model by fitting the model power spectrum to the EEG recordings. To this end, we address the task of parameter estimation in the models that are described by a set of stochastic ordinary or delay differential equations [2].
Motor system
Participants: Laurent Bougrain, Cecilia Lindig-Leon, Jimmy, Nex, Sébastien Rimbert.
In collaboration with Stéphanie Fleck (Univ. Lorraine), Nathalie Gayraud (Inria Sophia Antipolis) and Maureen Clerc (Inria Sophia Antipolis)
Incremental motor imagery learning for rehabilitation after stroke
After a stroke, Brain-Computer Interfaces (BCI) allows improving rehabilitation of the motor cortex to recover the autonomy of the patient. The design of BCIs has to be done with an in-depth analysis concerning user’s conditions during the learning of BCI. Since strokes affect mainly senior citizens, it is very important to guide the design of BCIs to make it usable. We propose to improve the experimental conditions through a new BCI protocol including an incremental motor imagery learning [21].
Motor neuroprostheses
We wrote a review that aims to position current neuroprosthetics research between reality and fiction, expectations of persons under a disability, fantasies of the augmented Man and scientific difficulties. Beyond the buzz effect to get the attention of the public and funders, and enthusiasm by journalists for novelty what are the expectations of potential users, the disappointments and the satisfactions of patients, how many persons are equipped, what are the price and the opportunities to use such devices outside of laboratories [5].
Classification of Motor patterns
In order to build systems that are able to detect several motor patterns, multiclass schemes need to be applied. We compared a series of multiclass approaches to assert the benefits of hierarchical classification. The compared methods are based on two effective techniques for MI-discrimination, namely, Common Spatial Patterns (CSP) and Riemannian geometry, for which the hierarchical and non-hierarchical approaches have been considered. We include the CSP by Joint Diagonalization method, which corresponds with a non-hierarchical approach; and its hierarchical counterpart, namely, Binary CSP. In addition, the non-hierarchical Minimum Distance to Riemannian Mean method (MDRM) is also evaluated, together with its analogous hierarchical approach; a contribution of the present work called Hierarchical MDRM algorithm (HMDRM). All these methods have been applied on dataset 2a of the BCI competition IV to facilitate their comparison. The highest accuracies were reached by the BCSP and HMDRM methods, confirming the effectiveness of hierarchical algorithms [7].
Discrete Motor Imageries for a Faster Detection
We are investigating differences between continuous MIs and discrete MIs corresponding to a 2s MI. Results show that both discrete and continuous MIs modulate ERD and ERS components. Both ERSs are different but ERDs are close in term of power of (de)synchronization. These results show that discrete motor imageries may be preferable for BCI systems design in order to faster detect MIs and reduce user fatigue. [8]
Pain under General Anaesthesia
Participants : Mariia Fedotenkova, Axel Hutt, Tamara Tošić
In collaboration with Peter beim Graben and James W. Sleigh.
Detection of EEG-signal Features for Pain under General Anaesthesia
Mariia Fedotenkova defended her thesis about extraction of multivariate components in brain signals obtained during general anesthesia. We studied analgesia effect of general anesthesia, more specifically, on patients reaction to nociceptive stimuli. We also study differences in the reaction between different anesthetic drugs. The study was conducted on a dataset consisting of 230 EEG signals: pre- and post-incision recordings obtained from 115 patients, who received desflurane and propofol. Combining features obtained with power spectral analysis and recurrence symbolic analysis [22], [6], [23], classification was carried out on a two-class problem, distinguishing between pre-/post-incision EEG signals, as well as between two different anesthetic drugs, desflurane and propofol [1].